(Sunys 2 PDF
(Sunys 2 PDF
50
50,39,39
100
PHILOSOPHY
Anaximander in Context
100
Promoting a new, broadly interdisciplinary horizon for future studies in early Greek
100
philosophy, Dirk L. Couprie, Robert Hahn, and Gerard Naddaf establish the cultural con-
100
text in which Anaximander’s thought developed and in which the origins of Greek phi-
RED
achievement, the authors call our attention to the historical, social, political, technological,
BLU
arguments that help to illuminate Anaximander’s thought, in particular the need to com-
Anaximander in Context
prehend the culture of Archaic Miletos, the strong role of architecture as an essential
75
element to understanding Anaximander (not a new view but one here carried further
50
than ever before), and the need not to dismiss Anaximander’s stranger views on astronomy
25
the Architects:The Contributions of Egyptian and Greek Architectural Technologies to the Origins
of Greek Philosophy, also published by SUNY Press. Gerard Naddaf is Associate Professor
50
and Chair of the Department of Philosophy at York University. He is the translator and
25
www.sunypress.edu
50
25
100
100
100
SUNY
100
10 9 8 7 6 5 4 3 2 1
Contents
List of Illustrations ix
List of Tables xiii
Introduction 1
Bibliography 255
ix
x LIST OF ILLUSTRATIONS
xiii
This page intentionally left blank.
Introduction
1
2 INTRODUCTION
and social) originated for the social group. Naddaf argues that Anaximander
was attempting to accomplish the same end. Anaximander’s cosmogonical
and cosmological perspectives were developed by the author in two recent
studies (1998c, 1–28; 2001, 5–21). He now attempts to put the anthropogonical
and politogonical aspects into perspective. To illustrate Anaximander’s new
revolutionary approach, Naddaf begins with a synopsis of the major Greek
mythical approaches to the origin of humanity. He then shows that for
Anaximander living beings emerge from a sort of primeval moisture or slime
activitated by the heat of the sun after the initial formation of the universe.
Naddaf argues that the doxographical evidence suggests that Anaximander
defended a doctrine of the transformation of species, and that climatic con-
ditions were behind the numerous changes or modifications in animals. In
conjunction with this, he argues that Anaximander recognized a connection
between his hypothesis of a drying earth and his zoogonical theories. More-
over, he shows that Anaximander surmised for valid reasons that humans did
not undergo a transformation completely similar to that of other animal
species. Naddaf contends that Anaximander’s account was so audacious that
it was not really surpassed until Jean-Baptiste de Lamarck proposed in 1809
the first coherent theory of the evolution of species.
Before reconstructing and examining Anaximander’s own account of the
development of society, Naddaf recalls how this development was perceived
before the advent of speculative thought. He then gives a historical analysis
of the primary causes that gave rise to the polis or city-state. The importance
of the polis resides in the fact that without it, Western philosophy and sci-
ence, and thus the new rational/naturalistic approach, would not have seen
the light of day. Naddaf pays particularly close attention to the polis of Miletus.
Although tradition situates the birth of philosophy in Miletus, scholars too
often have ignored its history or have limited it to a few cursory remarks. A
close analysis of the history of Miletos reveals some important and interesting
facts regarding the birth of philosophy. Naddaf shows that the long period of
extreme social unrest mentioned by Herodotus and others and situated dur-
ing Anaximander’s period is unfounded in light of other events. He also
analyzes the close military and economic ties that existed between the Ionians
in general and Miletus in particular and Egypt from the middle of the sev-
enth century B.C.E. to the time of Anaximander. In conjunction with this,
Naddaf traces the Greek belief that Egypt was both the oldest civilization
and the cradle of civilization to the time of Anaximander. Indeed, he argues
that it may have originated with Anaximander himself. Anaximander was no
armchair philosopher. He formulated his theory through investigation and
discovery and traveled extensively, notably to Egypt via Naucratis. In this
regard, he attempts to show that Egypt, or more precisely, the Nile Delta, was
in certain respects the center of the universe, for both humanity and civili-
Introduction 3
zation originated here. Naddaf believes that there is a good deal of circum-
stantial evidence for this, and he attempts to put this evidence into perspec-
tive, notably in his reconstruction of Anaximander’s famous map of the
oÎkouménh or inhabited world which, in the final analysis, was constructed
to show how and why civilization originated and propagated from Egypt.
Naddaf argues that some of the evidence here will corroborate Martin Bernal’s
claims regarding the relation between Greece and Egypt, albeit for different
reasons. It is all part of what one author has called “the Egyptian mirage in
ancient Greece.”
In his Proportions and Numbers in Anaximander and Early Greek Thought,
Hahn begins where Naddaf left off, namely, with the Egyptian connection. In
Anaximander and the Architects: The Contributions of Egyptian and Greek Archi-
tectural Technologies to the Origins of Greek Philosophy (2001), Hahn argued that
in the absence of any adequate astronomical theory or instruments, Anaximander
had imagined the shape and size of the cosmos by architectural techniques.
Anaximander appealed to these techniques because he imagined the structure
of the cosmos as “cosmic architecture.” In Anaximander and the Architects, Hahn
argued that the architectural techniques that Anaximander imported into his
own thought were inspired by the Greek architects building stone temples of
monumental proportions in his Ionian backyard. These techniques themselves,
however, were imported into Ionia from Egypt, from the middle of the seventh
century and onward, through the same avenues of transmissions that Naddaf
also identifies, and through the Milesian trading colony at Naucratis. In Hahn’s
monograph, he begins by framing the Egypt connection, focusing upon the
transmission of architectural forms and techniques, but this time he sets out the
specific monumental building projects and innovations introduced in Egypt
during the seventh and sixth centuries B.C.E. While the Greek architects (and
perhaps Anaximander and Thales themselves) had a chance to learn from the
Egyptian architects and their projects, the archaic architects of Ionia did not
simply copy Egyptian temples but rather made temples of a peculiarly Greek
type. Not only in form did the Ionian dipteral temples differ from those of their
Egyptian counterparts but also in the proportions and numbers. In order to
explain Anaximander’s selection of proportions and numbers, then, Hahn first
turns to reconstruct Anaximander’s cosmic architecture and then explores fur-
ther the proportions and numbers in the first great stone temples to Artemis
in Ephesos, to Hera in Samos, and Apollo in Didyma.
In Anaximander and the Architects, Hahn had shown that the architects
had imagined their enormous temples from more than one point of view, that
is, from plan or aerial view (that is, directly overhead) in addition to three-
dimensional views. Furthermore, he also made clear that the architect’s suc-
cess in monumental building rested on a design technique, on identifying a
module, and then on calculating each part of the temple as multiples or
4 INTRODUCTION
In Anaximander and the Architects, Hahn argued that the numbers surviv-
ing from tertiary reports concerning Anaximander’s heavenly wheels could be
accepted—“19” for the moon wheel, and “27” and “28” for the sun wheel.
Although the usual recitation in the scholarly literature has been to identify
the heavenly order with the geometrical series 9, 18, and 27, Hahn argued
that the numbers should be expressed differently. Following the architectural
analogy, the distance from the earth to the heavenly wheels is 9/10 earth
diameters to the stars, 18/19 to the moon, and 27/28 to the sun. In each
assignment of numbers, the first number is the distance to the inside of the
wheel, the second number is to the outside, and each wheel is one module in
diameter. Hahn argues, in agreement with Couprie, that the numbers repre-
sent distances to, not sizes of, the heavenly wheels. Since Anaximander’s
cosmos is calculated in terms of the module, Hahn argues in this new mono-
graph, that Anaximander’s heavenly wheels also should be expressed in terms
of the column-drum earth module. The relation between the diameter and
depth of Anaximander’s earth is 3 1. When the distances to the wheels are
expressed in modular formulas, they are 3 3 1 + 1 (9/10), 3 3 2
+ 1 (18/19), and 3 3 3 + 1 (27/28). Hahn shows that this formula was
precisely one in use by the architects in Miletus/Didyma, contemporaneous
with Anaximander. The evidence for this architectural design formula comes
from the recent reports of the archaeological excavators. Furthermore, when
Anaximander’s numbers are called upon to express not the distances to but
the sizes of the heavenly wheels, the numbers bear a remarkable likeness to
the proportions and numbers by which the temple architects grew the
“Kernbau,” the sacred built structure of the temple inside of the peristasis or
colonnade. Thus Hahn reaches the conclusion that Anaximander may well
have imagined—by analogy—the cosmos itself as a cosmic Kernbau. And this
means also that he had to adapt the geometrical techniques that produced the
rectangular architecture of the temple to the circular and cylindrical geometry
of the heavens.
In his The Discovery of Space: Anaximander’s Astronomy, Couprie argues
that Anaximander deserves to be seen as the first great astronomer in the
West because he was the first thinker to forge the unique conception of space
so central to our understanding of the universe. While some historians of
science regarded the doxographical reports on Anaximander’s astronomy to be
so bizarre that they hesitated even to reproduce them, Couprie argues that
the evidence, handled with sufficient patience and care, presents a coherent
image that agrees both with the texts and with our observations of the sky.
In order to grasp the startling novelty of Anaximander’s conception,
Couprie asks the reader to abandon all modern preconceptions. These
preconceptions include the suppositions that the earth is spherical, that it
rotates each day on its axis, and that it revolves in an orbit around the sun.
Instead, Couprie insists, we must begin again, this time by imagining the
6 INTRODUCTION
heavens from a flat earth, and we must have no other astronomical device at
our service other than a gnomon, that is, a stick set perpendicularly in the
sand. Anaximander’s picture of the cosmos, then, is contrasted with contem-
porary ones from Egypt and Assyria. In Egypt, the heavens were conceived
as an overarching goddess; in Assyria, the heavens were conceived as an
extended mantle. To distinguish Anaximander’s picture of the heavens from
his ancient contemporaries, Couprie devises a series of “exercises in ancient
astronomy,” in which he shows several instances of what he calls the “anach-
ronistic fallacy.” One of the exercises is an attempt to reconstruct Anaximander’s
map of the world. If we begin by assuming a flat earth and try to picture an
“equator” and a “tropics,” Couprie, following Heidel, points out that these
lines, familiar to us for a spherical earth, now appear at completely unex-
pected locations. In comparison with an Egyptian or a Babylonian map,
Anaximander’s map appears as a complete innovation; indeed, it marks the
origin of what we have come to call a “geographical map.”
In the core of his monograph, Couprie argues that Anaximander taught
three epoch-making astronomical theories: (1) that the celestial bodies make
full circles and thus also went underneath the earth; (2) that the earth hangs
free and unsupported in the center of the universe; and (3) that the celestial
bodies lie behind one another. He stresses how daring these innovative theo-
ries must have been in the esteem of Anaximander’s contemporaries.
Anaximander is shown to counter their anxious question of why the earth
does not fall by his famous argument ex principio sufficientis rationis, which
has played a role in philosophical discussions until Leibniz. We have to wait
until Newton’s laws of gravity for a better answer. Couprie shows that the
unexpected order in which Anaximander places the celestial bodies (first the
stars, then the moon, and finally the sun) is not so strange as it seems at first
sight. The much-discussed question of Anaximander’s numbers (9, 18, and
27, indicating the distances of the heavenly bodies) gets an elegant, simple
solution, as they are taken to mean “far,” “ farther,” and “farthest.”
Couprie argues that several reports on Anaximander’s universe might be
understood as instructions for drawing a map of the universe. He shows a
reconstruction of this map, on which Anaximander’s numbers stand for the
distances to the celestial bodies. Couprie has serious doubts about whether
Anaximander could have made a three-dimensional model. Nevertheless,
Couprie offers such a model that is shown to be in accordance both with the
doxographical reports and with observational data.
As a kind of appendix, Couprie puts forward an interpretation of some
pictures of the ceilings of Egyptian temples dating from the Ptolemaic pe-
riod. It concerns duplicate or triplicate representations of the goddess of the
Heaven, Nut, of which no satisfying interpretation has been given. Couprie
offers the hypothesis that this intriguing change in the representation of the
heavens can be seen as an instance of Anaximander’s influence.
Anthropogony and Politogony
in Anaximander of Miletus
Gerard Naddaf
PROLOGUE
9
10 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
the origins of humanity and society/civilization, they must be put into per-
spective. This will necessitate a certain amount of digression, but the great
Milesian will always be in the background. When we turn to Anaximander’s
own position on the origin of civilization, much of the focus will be on a
reconstruction of his famous map and how it explains, in light of the
doxographical and historical evidence, the real aim of his own historia.
According to W. A. Heidel, the aim of Anaximander’s book was “to
sketch the life-history of the cosmos from the moment of its emergence from
infinitude to the author’s own time.”1 This also is precisely what Hesiod
attempted in the Theogony. He sought to explain how Zeus established the
present order of things, natural and social. This is the aim of a cosmogonical
myth in general, and Anaximander is clearly attempting to accomplish the
same end. This is why he must begin with a cosmogony, and go on to an
anthropogony, and end up with a politogony.2 However, his approach is radi-
cally different, since his explanation is not only naturalistic, but he clearly and
distinctly separates all three developments. Meanwhile, Anaximander was no
armchair philosopher. He formulated his theory through investigation and
discovery; he traveled extensively, notably, it will be argued, to Egypt via
Naucratis. In this regard, I attempt to show that Egypt or, more precisely, the
Nile Delta, was in certain respects the center of the universe, that is, the
center before the shift, the shift to Miletus. I believe that there is a good deal
of circumstantial evidence for this, but the argument must be read as a whole.
Some of the evidence here will corroborate Martin Bernal’s claims regarding
the relation between Greece and Egypt, albeit for different reasons. It is all
part of what one author has called “the Egyptian mirage in ancient Greece.”
The fact that everything that gives birth is supplied with the food its
offspring needs is weighty testimony for this assertion that the earth
hereabouts gave birth to these men’s ancestors and ours. For by this sign
it can be seen clearly whether or not a woman has really given birth: she
is foisting off an infant not her own, if she does not have within her the
wellsprings of its nourishments. The earth here, our mother, offers pre-
cisely this as sufficient testimony that she has brought forth humans.
She first and she alone in that olden time bore food fit for humans,
wheat and barley, which are the finest and best nourishment for the
human race, because she really was the mother of this creature. And
such testimonies are to be taken more seriously on earth’s behalf than
a woman’s, inasmuch as earth does not mimic woman in conceiving and
generating, but woman earth. (trans. Paul Ryan)
eponymous ancestor who was both fully human and prestigious by birth (not
to mention the distinction of having given birth to humanity). From this
perspective, the present could find its justification in a genealogy that linked
it to the distant past.
An example of the second case is offered by the first couple: Deucalion and
Pyrrha. According to the legend, Deucalion was the son of Prometheus and
Pryneia, and Pyrrha was the daughter of Epimetheus and Pandora.11 A first
humanity originated from this primordial couple, however, they were subse-
quently destroyed by the Flood (Pindar’s Olympian Ode 9.42–53). It is uncer-
tain who caused the Flood, but Apollodorus (1.7.2) states that it was Zeus in
one of his notorious rages. Deucalion and Pyrrha, however, were warned in
advance by Prometheus.12 This warning allowed the couple to build the Greek
ark (or chest: lºrnax) in which to take refuge while the cataclysm wiped out
the rest of humanity. When the rain finally stopped and the couple touched
firm ground, Deucalion offered Zeus a sacrifice, and in return Zeus sent Hermes
to grant Deucalion a wish. Deucalion desired that there once again be men
(lao√). As a result, a new humanity originated from the stones (lø'V) that the
husband and wife threw over their shoulders. When the stones touched the
earth, those thrown by Deucalion became men, while those thrown by Pyrrha
became women.13 It is only later that Deucalion and Pyrrha sired their own
children, including Hellen, the ancestor of all the Greeks.
According to this myth, a new humanity was thus born or reborn from
death. Indeed, not only is the Greek ark (lºrnax) a source of life (or rebirth)
for humanity, but it also serves as a coffin (see LSJ). This is consistent with
the play on words or the etymological creation between løaV (stone) and
la¬V (people).14 Stones, seemingly inanimate objects, also are the source of
life and death for men. As Pindar so pertinently observes, Deucalion and
Pyrrha first gave birth to the “stone people” (l√qinon g¬non, Olympian Ode
9.45–46e; see also Apollodorus 1.7.2). Thus it is not surprising that it is
without sexual union that Deucalion engenders the Òndr'V and Pyrrha en-
genders the gunaƒk'V, that is, the same produces the same. Subsequently,
however, the couple will procreate to bring Hellen to the world. This signifies
that the existence of Pyrrha as a woman is finally assured.15 The idea that
many humanities succeeded each other following similar downfalls or cata-
clysms (floods) is certainly not absent in the new rational thought, although
it insists, as we shall see, on empirical observation to explain this succession.
living beings (not mentioned by the poets and/or in mythical accounts) is, as
in the case of his cosmology, the first naturalistic explanation in this domain.
As one might expect, his explanation is entirely consistent with his cosmo-
logical system. Indeed, the same natural processes are at work (DK 12A27).
Living beings emerge from a sort of primeval moisture or slime (÷x •gro£),
which is activated by the heat of the sun after the initial formation of the
universe.16 In sum, life results from the action of the hot and the dry on the
cold and the wet. Now although Anaximander clearly believed in a “sponta-
neous generation,” he did not believe, as did the poets, that humanity and
other animal species emerged “whole” from the earth. Indeed, there is an
astonishing consistency to his account. According to Aetius (12A30),
Anaximander argued that the first animals (tΩ prÍta z¸a) that arose from
primeval humidity (÷n •gr¸) were enclosed (or surrounded, p'ri'c¬m'na)
in a thorny bark (floioƒV ™kanqÔd'si),17 but that after they grew older and
matured, they emerged (™poba√n'in) on dry land, shed their thorny cover-
ings, and in a short time after (÷p ’ ıl√gon cr¬non) began a modified form
of existence (m'tabiÍnai), that is, adapted to their new environment.18 Al-
though it is clear that “all” living creatures arose from the primeval humidity,19
Aetius clearly is referring to potential land animals in his doxography (12A30).
It is therefore unclear if Anaximander thought that all living creatures were
originally covered in “thorny bark”(floioƒV ™kanqÔd'si) or only the first
potential “land animals.” Nor is it clear why Anaximander thought that the
first land animals were initially covered this way. The first thing that comes
to mind is that it afforded them some sort of protection. But protection from
what? Conche conjectures protection from some sort of marine animal such
as carnivorous fish.20 However, if Anaximander believed that all marine ani-
mals once had thorny encasements, even the carnivorous fish would have
been protected in a similar way. Indeed, Conche’s conjecture presupposes that
various species of animals already coexisted in the primeval marine environ-
ment (we may assume that Anaximander believed that the first creatures were
nourished by the primeval slime). Yet it is unclear (although commentators
tend to take it for granted) whether Anaximander thought that the various
“potential” land animal species already had different “forms” when they in-
habited the “marine environment.” What seems certain from the doxographical
evidence is that as some of these “thorny” creatures matured, they somehow
“migrated” to dry land. And once on dry land, the thorny skin was shed at
some point, and shortly after this their mode of living was modified accord-
ingly. Of course, it was only after the heat of the sun had evaporated enough
of the moisture for dry land to appear that the marine animals could in fact
migrate to it (which means that they were not immobile). In sum, the evi-
dence suggests (contra Barnes) that Anaximander recognized a connection
between his hypothesis of a drying earth and his zoogonical theories.21 Clearly
Gerard Naddaf 15
kind (÷x ™llo'idÍn z±wn), although he clearly has aquatic animals in mind,
since all land animals have an aquatic origin. Hippolytus confirms a relation
between humans and fish, although he says only that the human species was
originally similar to (parapl–sion) fish.27 But what does this mean? If we
did not have any other doxographical evidence, we simply would say that
humans had a marine existence before transforming into land animals. How-
ever, Censorius is much more explicit. He says that Anaximander thought
that humans were first formed in fish or fishlike creatures. Censorius accounts
for the way Anaximander perceived this. Originally the primordial sea (or the
primeval slime) must have secreted (after being acted on by the heat of the sun)
different kinds of embryonic life-forms, albeit not necessarily at the same time.28
Some of these embryos evolved into fish or fishlike creatures; others evolved
into land animals. Human embryos, on Censorius’ account, were, at some
point, somehow swallowed by fish or fishlike creatures but were able to sur-
vive like parasites. The human embryos were able, in time, to mature inside
of these creatures. When they did reach maturity, the fishlike creature erupted,
and men and women who were already able to fend for themselves emerged
and, one would presume, were able to procreate. Since we can assume from
Censorius’ account that human beings immediately took to dry land after
emerging from the fishlike creatures, it follows either that evolution was rapid
or that human embryos were secreted by the sea at a later stage. Plutarch
appears to confirm Censorius’ account when he states that Anaximander
(unlike the Syrians, who attributed a common parentage to fish and humans)
declares not only that humans and fish are from the same element, as the
Syrians do, but also that humans were first born in fish (÷n Îcq§sin ÷gg'n°sqai
t¿ prÍton ™nqrÔpouV). Indeed, he concurs with Censorius that humans
also were nourished in fish. What Plutarch adds is the type of fish in which
this occurred: the smooth dogfish (gal'o√), which is, like all sharks, a placen-
tal animal that gives birth to live young.29
This fish, according to Plutarch (On the Cleverness of Animals 982c; On
Affection for Offspring 494c), has a number of fascinating qualities, including
viviparous reproduction, nursing of the young within their own bodies, and
extruding the young and taking them back again.30 It is because humans were
cared for inside of sharks that they were (eventually) able to look after them-
selves and then (thnika£ta) come forth (÷kb›nai) and take to dry land
(g›V lab°sqai). Of course, Plutarch does not state that the dogfish ruptures,
as Censorius notes, after humans reach their maturity inside. The important
point to retain with regard to the testimonia of Censorius and Plutarch is that
Anaximander was genuinely concerned with accounting for the origin of
humans based on the fact that, unlike other land animals, they would be
unable to survive as a species without some initial help from Mother Nature.
Plutarch and Censorius may be just conjecturing on the source of their own
Gerard Naddaf 17
For ancient peoples, society comes into existence without a real past, in the
sense that it only reflects the result of a series of events that took place in illo
tempore, that is, before the “chronological” time of the people who narrate the
myth. Not only did these events unfold in a time and under conditions that
were very different from those experienced by the society narrating the myth,
but they also were due to the intervention of supernatural beings. It is for
these reasons that the members of archaic society considered their social
structure as defined once and for all.
In general, a cosmogonical myth is a traditional explanation of how the
world order (natural and social) originated for the social group. The world
order is seen as the result of the intervention of supernatural beings in another
world in a remote past that is different from the one in which the social group
lives. The world order generally is understood in terms of the “social reality” of
the other world. This “social reality” is perceived as the outcome of a series of
conflicts and/or agreements between gods. It is a sort of mirror in which the
society narrating the myth observes itself and measures its stability, and the
fact that this tradition is “performed” by the community only enhances its
persuasive power.
Ironically, the Homeric tradition has not bequeathed us a cosmogonical
myth strictly speaking, although it is clear that the social reality of Homeric
society, in particular the heroic aspect, is a mirror of divine or Olympian
society, which in turn mirrors the society (albeit with some confusion) in
which the community narrating the tradition resides. While the epics are not
without a good deal of historical authenticity (even a rationalist such as
Thucycides believed in the Trojan War), the poems represent an age in which
deities intervened openly in human life, in a way that later they did not. Not
18 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
that convey messages that the social group could have considered as having
been transmitted by its ancestors, in many respects it is advocating a new type
of social reform, a new type of general ™r't–.
In Works and Days as in Theogony, the kings are again at center stage; however,
the description offered by Hesiod in the former is radically different from the
latter. The story of the hawk and the nightingale (235–45) makes it clear that
the kings have a considerable, if not an absolute, amount of power over their
subjects, which they have no qualms about applying. Hesiod appears of the
opinion, as are most of us, that “absolute power” corrupts. The kings embody
©briV or violence, that is, the Homeric principle that might, or pure self-
interest, is right, as opposed to justice. Without justice, Hesiod believes that
people will devour themselves like wild animals, that there will be a sort of
Hobbesian state of nature—not unlike what preceded the reign of Zeus.
However, Hesiod is not easily intimidated, as the story of the hawk and
nightingale may leave one to believe.
Hesiod directly challenges the kings of Thespies with an astonishing
amount of “free speech.” The fact that the poem would have been “per-
formed” extensively throughout the Greek world only enhances this point.
The kings are unequivocally characterized as “greedy” and their verdicts as
“corrupt.” On three occasions in Works and Days, he describes them as “bribe-
eating” (dwrofºgoi, 39, 221, 264) and characterizes their sentences (d√kai)
as crooked or unjust (skolia√) (221, 250, 262). The kings also are contrasted
with the d›moV or people (261).
In Theogony, receiving gifts in exchange for delivering judgments is a
right of a mediator or king, and Hesiod painted there a rather flattering
picture of the custom. In Works and Days, Hesiod clearly is vexed by the
system of gifts. He doubts that the verdict or d√kh will be straight, and he
suggests that he has firsthand knowledge of this. Clearly, the lords appear
more interested in the gift than the sentence, and consequently they appear,
at least from Hesiod’s perspective, to announce a settlement so that the liti-
gants will surely return. In sum, the nobles are likely to be the only ones who
profit from this system. In Hesiod’s eyes, it must be replaced at any cost, for
clearly it has a “legal force.” What is worse, they see their d√kh as a question
of tim– (honor). Hesiod appears to be advocating nothing short of dispensing
with the kings, for they embody and indeed endorse the destructive eris
(competition or struggle). In fact, Hesiod appears convinced that the people
will pay for the arrogance of the nobles unless d√kh (justice/right) rather than
20 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
tim– becomes the central virtue (see Murray 1993, 61). This is clear from his
paradigm of the two cities in Works and Days: the city of d√kh and the city
of u
©briV or hubris (225–47). Hubris is responsible for famine, poverty, plague,
and so on (240ff.). However, Hesiod contends at 217–18 that d√kh eventually
will triumph over ©briV. D√kh will punish greedy men (220–23). The fact
that Hesiod employs d√kh twenty-one times from line 213 to 285 is indica-
tive of its importance. Hesiod appears to see justice as a “method of pro-
cedure” and appears to try to objectify the notion. The fact that D√kh is
personified and becomes the protector of society reinforces this conviction.
With the elimination of the kings or nobles, what remains—if not a call for
a written code of laws?39
Moreover, Pithekoussai provides the first datable evidence for the alphabet
and its early literary use in the form of a Homeric line inscribed on a cup (c.
750–700) (Boardman 1999, 168). Meanwhile, it appears that the search for
metals and luxury goods was to satisfy the aristocracy of Euboea. We know
that the two chief cities, Chalcis and Eretria, on the island of Euboea were
ruled by aristocratic communities that became engaged in the so-called
Lelantine war (c. 730–710). The war (of which Hesiod was well aware) was
fought for possession of the rich, arable Lelantine plain. According to
Thucydides (1.15.3), the war was exceptional, for it divided Greece into two
rival camps (including Miletus on the side of Eretria and Corinth on the side
of Chalcis). As Murray notes, the Lelantine war marked the end of an era,
because it was the last war fought in the old style between leading proponents
of that style. Of course, trade was not the only reason for colonization.45
Other reasons, and some would argue the primary ones, were land hunger,
famine, and political unrest. And in most cases, all three probably were con-
nected.46 Thus according to Murray (1993, 110), evidence exists of a large
surge in population growth in the eighth century, and this explains why land
was the chief colonizing factor. Since the traditional Greek custom of inher-
itance was to divide the land equally between all of the sons, the only way to
avoid acute social problems (and famine) was to organize settlement abroad—
generally on knowledge brought back by traders (111). The fact that Sparta
conquered and colonized its neighbors in the early seventh century must be
seen in this light (ibid.).
The colonists almost always retained strong ties with the mother city,
and each city had at least one founder (oÎkist–V) who acted as a leader and
who generally was from an aristocratic background (see Jeffrey 1976, 56–57).
His responsibilities included dividing the land into farms and setting aside
land for sanctuaries. Distribution would be by lot and on fair and equal terms.
Thus the first colonists could see themselves as a sort of landed aristocracy.47
Indeed, the first colonists often acquired privileges that the second wave to
the same location would not. Thus at Syracuse the first colonists were known
as Gamoroi, “those who shared the land.” The first colonists generally were
composed of small groups of 200 or less. This explains why some cities were
able to found several colonies within a single generation (Miletus being a
prime example). In order to be effective against, at times, far larger numbers
of hostile locals, one would assume that all men were of fighting age and
equipped accordingly (i.e., with heavy armor). This also would reinforce the
conviction that all of the founding colonists should have an equal voice in
political/community decisions.
Around the same time that law codes were being devised, another Greek
institution of revolutionary proportion arrived on the scene: the hoplite pha-
lanx. Hoplites were heavy-armed soldiers (including helmet, corslet, greave,
Gerard Naddaf 23
solid front round shield, and thrusting spear and sword) who fought in tight
formation in ranks that often were four to eight deep. It thus replaced the
individual champions of the Homeric tradition. Although order, discipline,
and courage were now the essential qualities of the fighting force—for break-
ing ranks could be devasting—the more hoplites, the more effective the force
would be. The rise of the hoplite phalanx usually is dated to around the mid-
seventh century B.C.E., although there is evidence for the invention of hoplite
tactics around 700 B.C.E.48 Indeed, the first colonists already may have sur-
mised something akin to this for mutual protection. What is certain is that
to create a force of several hundred (and eventually thousands) heavy-armed
hoplites necessitated both a large supply of metal and enough wealth and
expertise to be able to acquire and produce the heavy armor. As noted above,
the search for metal was one of the main catalysts behind the creation of
trading posts and eventually colonization. And since Odysseus makes it clear
that the activity of a sea-faring merchant was not an activity of high social
status (Odyssey 8.159), then while the aristocrates themselves may have ini-
tiated the impetus to trade and explore, they were not directly involved.
Merchants and traders were thus in a position to acquire a good deal of profit
(and thus wealth) in return for a great deal of risk in procuring the goods that
the aristocrats coveted, in particular, metal to create their weapons, armor, and
luxury goods.
In the Odyssey (17.383–38), Homer mentions itinerant craftsmen in the
service of a noble clientele. Now there is clear evidence that foreign craftsmen
were offered opportunity for migration, in particular, Phoenican metalwork-
ers (Burkert 1992, 23; Murray 1993, 82; Boardman 1999, 56–62). It also is
clear that Greek craftsmen were quick to adopt the techniques and create
their own masterpieces—albeit with strong orientalizing motives (Burkert
1992, 16). In Hesiod’s Works and Days (24–26), he attests to the spirit of
independence and competition that reigned in the Greek workshops.49 More-
over, the passage suggests, on the one hand, that already in this period, the
dhmiourgo√ had begun their sedentarization and grouping in urban centers
close to the merchants, shipowners, and other nonagricultural workers. On
the other hand, it suggests that the demystification of technical ability (in
Homer, any technical skill is the result of divine inspiration) also was under-
way. In sum, Hesiod’s famous passage strongly suggests that there is already
a new spirit of secular competition at work.
At the beginning of the seventh century, if not earlier, it is thus clear that
there were numerous city-states with a strong well-to-do nonaristocratic
component. It also is clear that this nonagricultural component was a primary
source of wealth for the city-state. At what point did this new group of
wealthy individuals demand equality with their aristocratic counterparts?
According to Aristotle (Politics 4.1297b), a shift in political power (and
24 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
d›moV. At least initially, it seems that the popular leaders came from the
aristocracy, and that it was therefore a “popular form” of government set up
against the aristocracy (Murray 1993, 139). Of course, the d›moV would have
been powerless without the hoplites. This also may explain why tyrants were
later replaced by hoplite-dominated governments.
Tyranny, of course, was not always a fact of life before the codification
of law. There was an attempt at tyranny at Athens in 630 by an aristocrat,
Cylon. However, he had little, if any, popular support. The reply to this,
however, was the famous law code by Draco in 620. This suggests that already
there was strong discontent in Athens. However, Draco’s laws did little to
curb the infighting, which explains why Solon (c. 640–560) was elected as
archon and arbitrator in 594 ( Jeffrey 1977, 90). Solon could have made
himself tyrant, but he preferred to persuade the two sides, the people and
those who had the power, to agree to a fair compromise (the lawgiver in his
role as arbitrator). While those who had the power kept their property, the
people were given the dignity that was their due, including the cancellation
of all debts and the abolition of enslavement for debt. All Athenians also were
now divided into four classes that were determined by “wealth” and to each
of these corresponded their share of political power (to the top classes went
the top offices, to the lowest, only membership in the assembly). Clearly,
wealth rather than birth was the criterion for political honor, and Solon
himself noted that agriculture, trade, technology, and even the “intelligentsia”
(poets, seers, and doctors) were acceptable ways of acquiring wealth (fragment
13.43–17). This explains why two craftsmen were among the ten archons ap-
pointed in 580 B.C.E. (Murray 1993, 199). Moreover, Solon appealed to no
other power than his own sense of what was just: “I wrote down the laws alike
for base and noble, fitting straight judgements to each” (frag 36.18–20 = 24d).53
Solon presents the benefits and sanctions of law as human, and since the laws
were written down, it meant that they were available for all to see. Also, since
the laws were composed in poetic form, they could be easily memorized and
thus recited in public both by Solon and others (Murray 1993, 182–83; Naddaf
2000, 347–49). Much of Solon’s political reforms did, however, fail, and there
was a period of unrest that lasted until Peisistratus successfully seized power as
a tyrant in 546. But ironically it was during the Peisistratus tyranny that Athe-
nians enjoyed a period of stability and great prosperity manifested in the con-
struction of public works and the creation of national festivals and games, and
thus a developed sense of national unity, local pride, and individual diginity.
Nonetheless, when tyranny did end in 510, there would be no return to the old-
style politics. The stage was now set for the truly democratic institutions of
Cleisthenes in which the people would have a share of everything.
What about Miletus, the birthplace of Anaximander? Miletus was founded
around 1050 B.C.E. by Neleus, the youngest son of Athenian King Codrus
26 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
(Herodotus 1.47; 5.65; 9.97).54 Neleus and his men killed the natives (Carians)
and took their women as wives (Herodotus 1.146; Huxley 1966, 32). We
know that the Milesians were quite involved in the famous Lelantine war (c.
730–680).55 They sided with Eretria against Chalcis, something the Eretrians
never forgot (Herodotus 5.99). It may have been at Al Mina (see above) that
the Milesians first became involved in the Euboean rivalry. Given that the
Lelantine war was fought essentially between aristocrats, this strongly sug-
gests that the Neleids were still firmly in control at the time.56 However, there
is some evidence that it was around the time of the Lelantine war that the
kingship in Miletus ended. It is associated with a struggle between two Neleids:
Leodamus and Amphitres.57 Both were engaged in wars at the time: Amphitres
against Melos and Leodamus against Carystus in southern Euboea. This
suggests that it was actually Leodamus who led the Milesians forces during
the Lelantine war (Huxley 1966, 50; Jeffrey 1977, 210). Amphitres mean-
while lost his campaign, while Leodamus won. Nevertheless, Amphitres later
murdered Leodamus and seized power by force. But the exiled sons and
friends of Leodamus returned, and when the two sides clashed, the sons of
Leodamus were victorious and killed Amphitres. The tyranny of Amphitres
was thus short lived. Indeed, after peace was restored, it seems that the d›moV
or citizens elected a lawgiver (aÎsumn–thV) or a “temporary dictator” called
Epimenes. He was given the power to put to death the sons of Amphitres
(and their allies) and seize their property. Although the election of Epimenes
effectively ended Neleid kingship in Miletus, the Neleids remained, of course,
a powerful group, a sort of aristocratic oligarchy.58 It is possible that they
continued to dominate the political scene at Miletus until their power was
effectively curtailed by Thrasybulus around 630 B.C.E. Why this date? Pseudo-
Skymnos (Geographical Description 986–97) says that Sinope was refounded
by exiles from Miletus, and scholars date this around 630 B.C.E.59 I suggest
that the exiles were Neleid aristocrats who met the same fate as their Corinthian
counterparts, the Bacchaids. This may explain the close friendship between
Periander and Thrasybulus. Indeed, the way Thrasybulus seized power is
somewhat analogous to the way Cypselus did: both were initially in a position
of authority within their respective oligarchies. In the case of Thrasybulus,
Aristotle is quite explicit. One reason oligarchies were initially capable of produc-
ing tyrants, he says, was because they gave to one official supreme authority over
all other offices. He cites as an example Miletus with its prytanis (or president)
and states that this is how Thrasybulus became tyrant (Politics 5. 1305a16–18;
Jeffrey 1977, 210; Huxley 1966, 50–51 Gorman 2001, 99–101). Since Thrasybulus
obviously needed the support of the army and the people to become a tyrant, the
prytanis must have included the office of polemarchos or military leader.60
We do not know what, if anything, Thrasybulus promised the people and
the army for their support. He may have promised them a share in govern-
Gerard Naddaf 27
ment, as Cypselus had done,61 and may have divided up the estates of the
exiles. Again, like Cypselus, he may have revised and codified the existing
laws, thus ending the arbitrary interpretations of the Neleids. Was it
Thrasybulus who initiated a sense of isonomia or “equal rights” in Miletus,
something that was already more widespread in their colonies! It seems that
Thrasybulus is clearly a much more important figure in the history of the
great city then is usually recognized. Murray may be right to contend that at
the end of the seventh century, Thrasybulus “brought Miletus to the height
of her power” (Murray 1993, 138; see also A. Andrewes 1956, 118). This did
not occur before Miletus was confronted with the formidable power of its
Lydian neighbor. However, if Miletus was able to thwart its powerful neigh-
bor, it is because it already possessed a strong army and, more important, a
mighty navy. This in turn suggests, indeed corroborates, that Miletus was an
economic powerhouse even before Thrasybulus seized power.
Between the eighth and sixth centuries, Miletus established numerous
colonies, particularly in the area of the Black Sea (see Boardman 1999, 238ff.).
Pliny states that Miletus was responsible for setting up over ninety colonies
(Natural History 5.31.112; see also Strabo 14.1.6). Although the search for
land was certainly a motive, it seems that trade was the primary factor. Of
course, colonization is not a necessary condition for trade. Given the prox-
imity of Miletus to its eastern neighbors (e.g., it was less than 100 kilometers
from Sardes, the capital of Lydia), trade with the interior by caravan also
could be profitable and of course dangerous.
Around 687 B.C.E., Gyges, a member of the Lydian family of the
Mermnadae, overthrew the last Heraclid king of Lydia, Candaules, and married
his wife, Toudo (Herodotus 1.12). Shortly after this, Gyges (whose reign is
described by his contemporary, poet-soldier Archilochus, as a tyranny) at-
tacked Miletus and other Ionian cities (Herodotus 1.15). Although he sacked
Colophon, he was unsuccessful against Miletus, which must have been
sufficiently rich and menacing at the time to warrant an attack. Herodotus
contends that this was the only attack on Miletus by Gyges, who was killed
in a battle against the invading Cimmerians around 652.
It seems that the relation between the Lydians and the Milesians was
somewhat ambiguous. Gyges allowed the Milesians to establish a colony at
Abydus on the Troad (Strabo 590). Of course, it may have been established
as a mercenary colony to help thwart the invading Cimmerians (Huxley 1966,
63); it certainly had a strategic location at the head of the Hellespont. It
seems that the Ionians quickly gained a reputation as mercenaries. Herodotus
(2.154) informs us that it was bronze-clad Ionian (including Milesian) and
Carian mercenary soldiers who helped the first Psammetichus (664–610) win
back his throne in Egypt. It has been conjectured that they may have been
sent by Gyges, who joined Psammetichus in opposition to the Assryians
28 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
(Huxley 1966, 72). These mercenaries were later rewarded with strips of land
in Egypt (Herodotus 2.154). In any event, this means that Greeks were in
Egypt by 664 B.C.E., and they wasted little time establishing a thriving trade
with Egypt. Herodotus (4.152) mentions the Samian merchant, Kolaios, who
was sailing to Egypt (c. 638) on one of his regular runs when he was blown
off course.
Gyges meanwhile was succeeded by his son, Ardys, who also invaded
Miletus—albeit again unsuccessfully—after the Cimmerians retreated, around
640 (Herodotus 1.15; Huxley 1966, 75). Herodotus seems to imply that
Ardys’ invasion was short lived, although his reign lasted for some time (c.
652–629).62 If Miletus was able to resist the invasions and threats of Gyges,
Ardys, and the Cimmerians, it was because it had both a powerful fleet and
a formidable army of hoplites that constituted, for the most part (in addition
to landowners), a wealthy class of merchants, shipowners, craftsmen, and the
like that continued to grow and prosper during this period. Ardys, mean-
while, was succeeded by his son, Sadyattes (c. 629–617). Sadyattes renewed
the war against the Ionians and, in particular, Miletus. Herodotus (1.17–19)
informs us that the territory of Miletus was ravaged for eleven consecutive
years: six during the reign of Sadyattes and five during the reign of his son,
Alyattes (c. 617–560), that is, 623–612. Each year the Lydians would sys-
tematically destroy the harvest.63 However, Miletus was well fortified, and it
also controlled the seas; grain could thus be imported from Egypt and else-
where. Consequently, the Lydian strategy failed. This occurred when Miletus
was under the tyranny of Thrasybulus (Herodotus 1.17–22). When Thrasybulus
did come to an agreement with Alyattes, the terms were very favorable to
Miletus. Indeed, they became friends and allies, and Miletus, contrary to many
of the other rival Ionian cities on the coast, maintained its independence.
The relation with Egypt must have been equally comfortable. Herodotus
(1921, 2.159) states that Egyptian pharaoh Neco (610–595) offered the ar-
mor he wore in his Syrian campaign to Apollo at Didyma in Milesia about
608 (Boardman 1999, 115). This suggests that Thrasybulus and Neco may
have been allies. Indeed, if the famous trading post at Naucratis in the Nile
Delta was founded around 620, then it also would have been under Thrasybulus
that the Milesians first settled there (see below). All of this adds force to
Murray’s contention that Thrasybulus “brought Miletus to the height of her
power” (1993, 138).
Was Thrasybulus an enlightened despot or a ruthless dictator? Herodotus
(5.92) contends that an initially mild Periander of Corinth began his ruth-
lessness after seeking and receiving advice from Thrasybulus.64 To the ques-
tion, what was the safest way to ensure control of affairs and best govern the
city?, the reply was to remove, from time to time, the preeminent citizens!
Aristotle reverses this and states that Thrasybulus sought the advice from
Gerard Naddaf 29
compromise as occurring during the reign of Croseus (c. 560–546), the son
and successor of Alyattes, and thus sees Miletus as flourishing as never before
under him (1988, 160).71 Huxley, for his part, argues that this period of
exceptional prosperity occurred under Persian rule (1966, 145).72 This makes
perfectly good sense, especially if we consider that Miletus was the only
Ionian city with whom Cyrus was willing to make a treaty (Herodotus 1.141),
and that the whole empire was now open to trade.73 The fact remains, how-
ever, that the Persians would only back a strong tyranny in Ionia, and this
included Miletus. Aristagoras (c. 513) makes it clear that if the tyrants do not
band together, they will all be overthrown by restless popular parties and
choose democracy in their respective cities (Herodotus 4.127–38).74 Since the
Ionian city-states fell to the Persians between 546 (fall of Lydia) and 539 (fall
of Babylonia), and since Cyrus clearly would not have tolerated the civil strife
of the intensity that many scholars associate with the period, this means that
the fraction fighting in question must have occurred prior to 540. Indeed,
given the Persian preference for tyranny, the commission that the Parians are
purported to have chosen among the Milesians itself would have had to have
chosen a tyrant among its members.75 Now this strongly suggests that the
civil unrest would have occurred during the reign of Croseus (560–546) and/
or earlier. But is there any reason to believe that Croseus, who had a xenia
with Miletus, would have tolerated the intense civil unrest any more than
Cyrus? In this case, perhaps Grant is correct that the Parian compromise
occurred during the reign of Croseus. This entails, in turn, that the intense
unrest occurred during the reign of Lydian King Alyattes. But in this case we
must ask why Alyattes would not have seized the occasion to conquer Miletus
and/or side with a tyrant to ensure stability in the region. Clearly, this was
the case after the Parian intervention, whenever it occurred. Of course, the
tyrant may have been elected, albeit from a limited group of individuals. But
was this any different from the tyranny under Thrasybulus? Given the cir-
cumstances and the particulars, it seems that there was never a period in
which Miletus did not prosper economically, and when strife did occur be-
tween factions, it seems that it did not last long. While there is no reason to
disbelieve that the strife was not at times intense, even as intense as Plutarch
describes, there is nothing to say that it was widespread to the point that
Miletus lapsed into utter chaos during a short period of time, let alone a
prolonged period of time. Clearly, there was a group, perhaps a very large
group, that occupied a position between the two extremes (and it may or may
not have had the support of the army and/or navy). As Phocylides, the great
Milesian poet and contemporary of Anaximander, explains: “Many things are
good for men who are in the just middle. In the city, I want to be a man of
the center” (Phocylides frag. 12 ed. Bergk; see Huxley 1966, 80, n. 46, 164;
West 1978, 166; in fragment 5, he expresses the importance of a small, well-
Gerard Naddaf 31
While Atlas is supporting the dome of the earth in the West, Prometheus is
represented as bound to a column in the East. Although the second column
clearly is taken from Hesiod’s Theogony (522), the rest of the cup decoration,
as a number of recent scholars have pointed out, appears to have been
influenced by the theories and teachings of Anaximander.86 Of course,
Anaximander hardly transported his map of the inhabited world on a stone
column, but the Spartan artist’s rendition would suggest a correlation be-
tween the two. However, are we to envision the famous map of the oÎkoum°nh
as round? What exactly was on the map? Is there a correlation between the
construction of the map and the contention that Anaximander invented or
introduced the sundial? And, more important, I think, what is the purpose of
the map? Let us begin with the shape of the map.
According to Herodotus, up until his time mapmakers generally depicted
the earth (g›) as perfectly circular with Ocean running like a river round it
and with Europe and Asia of equal size (4.36; although at 4.41 and 2.16 he
suggests that Asia, Europe, and Libya were perceived as having the same
size). In sum, the “father of history” gives the impression that the dominant
picture in his time was that of a circular earth surrounded by the river Oceanus.
However, Herodotus believes that this picture is clearly based on legend and
that “there is nothing to prove this” (4.8). The dominant picture is, of course,
Homeric,87 and this may be what prompts Herodotus to ridicule his prede-
cessors: “the absurdity of all the mapmakers” (4.36). Aristotle seems to be
making a similar point when he equally ridicules contemporary mapmakers
for depicting the inhabited earth (t‹n oÎkoum°nhn) as round. He also argues
that if it were not for the sea (and he bases this on observation), one could
travel completely round the earth (Meteorology 2.362b12). Of course, Aristotle
is imagining not a disk-shaped earth surrounded by the sea (or ocean) but a
spherical earth—albeit with a land/sea ratio of 5 to 3 (Meteorology 2.362b20–
25). On the other hand, Strabo seems to endorse the Homeric picture when
he argues that observation and experience clearly suggest that the inhabited
earth (≠ oÎkoum°nh) is an island and that the sea surrounding it is called
‘Oceanus’ (1.1.3–9). Agathemerus (3rd century C.E.) seems to add more pre-
cision to the picture of the map when he says that the ancients drew the
inhabited earth (t‹n oÎkoum°nhn) as circular (strogg§lhn), with Greece in
the center and Delphi at its center as the world’s navel (Geography 1.1.2=DK
68B15).
These references (but more importantly, Herodotus’) appear to suggest
that the most significant characteristic of early maps in general, and of
Anaximander’s in particular, is that they were circular. And there is virtual
unanimity among contemporary scholars on this point. However, a few re-
marks are in order. First, just what is Herodotus ridiculing here? The accent
seems to be on the notion of the legendary encircling river and on depicting
Gerard Naddaf 35
Europe and Asia of equal size. In sum, Herodotus is shocked by the “radical
rationalism” of the authors of these maps.88 Given Anaximander’s propensity
for perfect symmetery—the most perfect example of which is his cosmologi-
cal model—it is fair to say that the great Milesian phusikos was the inspiration
behind the early maps to which Herodotus refers in this famous passage. Of
course, we may wonder why the great rationalist would have argued that the
legendary river, Oceanus, encircled the earth. Clearly, Anaximander did not
believe in a legendary river, just as he did not believe that thunder and
lightning were Zeus’ prerogatives.89 As in the case of Strabo, observation,
experience, and legend probably would have led him to the conclusion that
the earth was surrounded by water. In fact, the earth, for Anaximander, was
initially covered with water—a point, in this context, that scholars have
ignored.90
Some scholars appear convinced that Delphi must be represented the
center of Anaximander’s map, since it was considered as the earth’s navel
(ımfal¿V g›V).91 I find it unconvincing that the great phusiologos would
succumb to popular belief any more than Herodotus or Xenophanes. Cer-
tainly he would have been aware that other civilizations (notably Egypt and
Babylon) claimed as much. More importantly, in Anaximander’s time, the
ımfal¿V g›V was probably the oracle of Apollo at Didyma near Miletus
rather than the oracle of Apollo at Delphi.92 Agathemerus may have been
thinking of maps from a later period (e.g., the same maps to which Aristotle
refers), or more “Hellenocentric” versions of similar maps. In sum, I would
suggest a more practical (and inspirational) point for the center of the map:
Miletus itself—although a good case also may be made for the Nile Delta as
we will see below. Indeed, according to Herodotus (1.170), when Thales,
following the defeat of the Ionians, suggested that the Ionians set up a com-
mon center of government at Teos, Teos was chosen for “practical reasons,”
not for “religious reasons.”
Furthermore, in the famous passage at 4.36, Herodotus states that it was
the g› or earth in general that his predecessors depicted as round, not the
oÎkoum°nh or inhabited earth.93 This comment may explain why Herodotus
adds that Europe and Asia, that is, the oÎkoum°nh, strictly speaking, are
depicted as being of “equal size” on the same maps—albeit much wider than
taller. Given Anaximander’s penchant for perfect symmetry, as noted above,
it seems quite plausible that Herodotus is indeed referring to Anaximander’s
map (or maps modeled on it). It is even more tempting, however, to take as
a reference to Anaximander Herodotus’ remark at 2.16 that the Ionians argue
that the earth consists of three parts: Asia, Europe, and Libya. Given
Anaximander’s fondness for the number three (the most perfect example of
which is, again, his cosmological model), it would appear to be his represen-
tation. Are we to imagine that the three were depicted as being of “equal”
36 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
size? At 4.41, Herodotus says that the three differ greatly in size, but since
Herodotus clearly is speaking for himself in this passage, there is nothing to
indicate that the reference at 2.16 is not to three continents of equal size.
Whatever the case, there is no reason to see the reference to two and three
continents, respectively, as being mutually exclusive. The reference to two
continents of equal size may be one to a map with a more precise frame, a
frame based on the use of the sundial and indicating the equator and the
tropics. I will return to this point shortly. Meanwhile, the division of the three
continents in Anaximander’s time would have been by rivers: the Nile in the
south, dividing Libya and Asia, and either the Phasis or Tanais (= Don) in
the north, dividing Europe and Asia.94 Since Anaximander believed that the
earth was encircled by the ocean, the exterior ocean would have been seen as
the source of the two rivers, canals, so to speak, carrying water to the more
centralized Mediterranean and Euxine or Black seas.95 But the Nile and
Egypt clearly have a special status for both Herodotus and his Ionian prede-
cessors, which merits a closer examination.
At 2.15, Herodotus states that the Ionians maintain that Egypt proper
is confined to the Nile Delta. Indeed, while the Nile is the boundary between
Asia and Libya, the Delta is seen as a separate piece of land (2.16). In the
passage that precedes this, Herodotus claims that the Egyptians believe that
they are the oldest race on earth (2.15; see also 2.1), and that they came into
being at the same time as the Delta (2.15).96 The rich alluvial soil of the Delta
enables the Egyptians to get their harvests with less labor than any other
people (2.14; see also Diodorus 1.34). Before giving his own opinion on the
subject, Herodotus notes that he is surprised that Ionians claim that the earth
consists of three parts, Europe, Asia, and Libya, when they should clearly
count the Egyptian Delta as a separate and fourth tract of land (2.16).
Herodotus’ own opinion (2.17) is unimportant here; what is important is
what the Ionians claim according to him (2.16), and what they clearly claim
is that Egypt is both the logical and chronological departure point/center for
humanity (2.15). Now there was an “almost” universal opinion among the
ancients that Egypt was the oldest civilization and the cradle of civilization.97
But with whom did this opinion/theory originate, and why?
There is little, if any, indication in Homer of the fabulous past of Egypt.98
Although it is rich in promises, Homer’s Egypt is not easily accessible.99 So
when did Egypt become accessible? The Greeks began to inhabit Egypt, as
we saw above, during the reign of Psammetichus I (664–610).100 Their resi-
dence in Egypt began after Psammetichus awarded a number of Greek
mercenaries strips of land. But it intensified in the late seventh century with
the foundation of the emporion or trading post of Naucratis, which is situated
about fifty miles inland on the Canopic branch of the Nile and thus only ten
miles or so from the royal capital of Sais (the capital of the XXVI Dynasty,
Gerard Naddaf 37
664–525) and seventy-five miles or so from the great pyramids of Giza (see
Herodotus 2.178–79). It has been argued that Naucratis initially was founded
by the Milesians (Conche 1991, 29, note 9). At the very least, it had a
Milesian quarter, as Herodotus clearly notes (2.178–79). Such a quarter is
suggested by a large, independent sanctuary that dates to the early years of
the town (Boardman 1999, 130; Gorman 2001, 56–58). Although Naucratis
acquired the status of a fully fledged polis, it was ultimately under the control
of the pharaoh. The importance of Naucratis in Egyptian history from the
time of Psammetichus cannot be exaggerated. It was, in fact, the chief port
of Egypt until the foundation of Alexandria, and it was not much less im-
portant than Alexandria in its own age, due in particular to the Philhellenic
pharaoh, Amasis (570–26).101 Moreover, since the Saite dynasty relied heavily
on mercenaries,102 it was only with the Persian conquest of Egypt by Cambyses
in 525 that the presence of Greek (and Carian) mercenaries ceased to over-
shadow the country (see Herodotus 3.12). Indeed, the Persian invasion had
an adverse effect on Naucratis itself. At one point, Darius suppressed the
privileged commercial relations with the city, and the archaeological evidence
seems to confirm this.103 As one recent scholar put it, the “heyday” of Naucratis
must have been before the Persian invasion of 525.104 Meanwhile, it is quite
possible that commerical (and touristic) relations with other areas of Egypt
began to develop after the foundation of Naucratis. The Greek influence
would have made it considerably easier for other Greeks to travel (not to
mention that the mercenaries themselves came from the four corners of
Greece), and thus for stories of a fabulous civilization to spread rapidly. It is
certainly not by chance, as Hurwit notes, that the colossal stone temples of
Artemis at Ephesus and the third Heraion of Samos with their grandiose
multiplication of columns date to the second quarter of the sixth century, at
a time when Rhoikos, the famous architect of the Samian temple, was visiting
Naucratis.105
Whether or not it was the Egyptians themselves who initiated the claim
that they were the oldest race on earth, and that they came into being at the
same time as the Delta, as Herodotus seems to claim, the fact remains that
the Greeks “quickly” believed this was the case. Why? Clearly, the sacred
colossal stone monuments, veritable museums of the past if there ever were,
with their indestructible archives (including a list of dynasties), certainly
conveyed the idea that humanity was much older than previously believed,
and at least considerably older than the genealogies of the oral tradition
claimed. Indeed, exposure to the Egyptian past very well may have incited a
whole new breed of individuals, including Anaximander, to rethink the origin
and development of humanity and civilization. In fact, Herodotus conjectured
that given the alluvial origin of the Nile Delta, it would have taken from
10,000 to 20,000 years to reach its present formation (2.11). This was ample
38 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
evidence for Herodotus that the Egyptians were indeed as old as they claimed
(see 2.142–145). In conjunction with this, the regularity of the Nile floods—
analogous to the regularity of the movements of the heavenly bodies them-
selves—and the subsequent annual renewal of the rich alluvial soil certainly
would explain how Egypt escaped the great cataclysms of the past, real or
imagined. Indeed, if the story of the flood of Deucalion were true—and most
Greeks, including the likes of Thucycides (1.3), Plato (Timaeus 22e–23d),
and Aristotle (Meteorology 352a30), thought it was—then the story would
have to be either reconsidered or the event considerably older than originally
thought.106 Alternatively, Egypt, that is, the Nile valley and Delta, escaped the
cataclysm that supposedly destroyed humanity.107 There were other ways or
reasons, that is, more “rational” ways or reasons of conceiving of, or postulat-
ing, a flood or something analogous. Herodotus (2.12; 13) concluded that the
earth was once covered by the sea from the observation of seashells on the
hills of the Nile Delta, and from the fact that salt exudes from the soil.
Whether or not he thought that this was a cyclical occurrence is unclear.
What is clear is that, according to Herodotus, geography/geology and history
are closely connected.
Herodotus informs us (2.143) that Hecataeus of Miletus (560–490) con-
jectured that Egypt had existed for at least 11,000 years after the start of the
records he was shown by the Egyptian priests in Thebes. Since Arrian in-
forms us that Hecataeus, in his History of Egypt, believed that the Delta was
formed by the continual deposit of silt, he probably assumed that the Nile
culture was even older and originated in the Delta.108 In sum, as in the case
of Herodotus, for Hecataeus, geography/geology and history are closely con-
nected. Given that this connection is similar to that made by Herodotus, and
given that Hecataeus is the source or inspiration behind Herodotus’ own
encounter with the Egyptian priest, it is reasonable to assume that Hecataeus
also is the source (or inspiration) behind Herodotus’ geological observations
about the Nile.
The early Ionian phusikoi also were interested in geology and cyclical
occurrences. Xenophanes argued that humanity (and life in general) emerged
from a sort of slime, that is, a combination of earth and water (DK 21B29,
33), and that it was periodically destroyed (DK 21A33). He based his theory
on the observation of various kinds of fossils (fish, plants, shells) in different
locations, including Syracuse, Paros, and Malta (DK 21A33). This was clear
evidence that the sea once covered what is now dry land. Did Xenophanes
and Hecataeus and, by implication, Herodotus have a common source, namely,
Anaximander of Miletus?109
Diogenes Laertius (on the testimony of Theophrastus) informs us that
Xenophanes was an auditor of Anaximander (Lives 9.21 = DK 21A1) and
gives his floruit (probably on the authority of Apollodorus) as the 60th
Gerard Naddaf 39
Olympiade (540–37), which suggests that he was born around 575. In his
autobiographical verses of fragment 8, Xenophanes informs us that he was
alive and writing at age ninety-two. There is a good deal of consensus that
Xenophanes lived from 575–475.110 Given the consensus on Anaximander’s
dates (610–540), if we assume that Xenophanes left his hometown of Colo-
phon after Cyrus the Mede’s conquest of Lydia and thus the reign of Croesus
in 546 (Colophon fell to Harpagus shortly after), then he would have been
around thirty at the time and Anaximander around sixty-five.111 Considering
the short distance between Colophon and Miletus, the facility of travel by
sea, and the reputation of Miletus as an intellectual center, Xenophanes could
have heard about Anaximander’s investigations and decided to attend his
private and/or public lectures (or even initially have “read” his famous book),
view his famous maps, and so on, somewhere between, say, 556 and 546.
Since Anaximander appears to have been a well-traveled man, this may have
incited Xenophanes to do as much. This is where he may have heard of
“cultural relativism” for the first time, as well as the relation between geology/
geography and history. We do not know where Xenophanes first traveled after
leaving Colophon for western Greece. He certainly was interested in the
origins and development of civilization and the arts that foster it (see below),
and his approach seems to be both rational and secular. Although none of the
references to fossils and seashells appear to be in Egypt (that is, unless the
reference to Paros, is Paros in Egypt [Marcovich 1959, 121]), his reference to
the Ethiopians as portraying their gods as flat-nosed and black (DK 21B16)
suggests if not a visit at least some familiarity with the culture.112 Indeed,
when Hecataeus states at the opening of his Genealogies that what the Greeks
believe is silly (of course, his criticism of anthropomorphism of the poets at
DK 21B11 also is in order here), he very well may be referring to Xenophanes’
contention concerning how the various peoples portray their respective gods.
Since Egypt and thus Naucratis did not fall to the Persians until 525, given
the reputation of Naucratis as a cosmopolitan intellectual center until 525,
then if Xenophanes did visit Egypt, it probably would have been before 525
(one exile at the hands of the Medes being enough!).113
Heidel puts the birth of Hecataeus at around 560 B.C.e.114 This strikes
me as being entirely plausible if one considers that Hecataeus probably was
an “elder statesman” during the Ionian revolt of 499, which he initially op-
posed. Now if Anaximander did live until at least 540 B.C.E. (and there is
nothing to indicate the contrary), Hecataeus also very well may have been a
young pupil/auditor of Anaximander (see Hurwit 1985, 321). Although he
generally is supposed to have visited Egypt during the reign (521–486) of
Darius, who was very favorably disposed toward the Egyptians and their
sancturaries after the lawlessness of his predecessor Cambyses (see Diodorus
1.95.5), Hecataeus also may have visited Egypt prior to this.115 Whatever the
40 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
case, that Hecataeus was a source and an inspiration for Herodotus is uncon-
tested by scholars. However, if Xenophanes and Hecataeus drew their inspi-
ration from Anaximander, then clearly the great Milesian phusikos also was
keenly interested in “chronology and geography,” although these were not, in
my view, his main interests, as Heidel contends.116 Was Egypt the source of
his own observations? Given the importance of Naucratis as a cosmopolitan
intellectual center with a Milesian quarter, and given that Anaximander (610–
540) had a reputation as both a traveler and a geographer, it would appear
strange if he had not visited the great country (few doubt that his friend
Thales did). Moreover, as noted above, Naucratis probably reached its zenith
prior to the Persian invasion of 525, and thus during Anaximander’s lifetime.
Was it a visit to Egypt that initiated his book on nature? Was Egypt the
catalyst?
Scholars tend to associate Anaximander solely, or almost solely, with cos-
mological speculation. But Strabo is quite emphatic that Anaximander the
philosopher (and his fellow citizen Hecataeus) were quite concerned with the
science of geography (1.1), and he goes on to say that this also was the opinion
of Eratosthenes (1.11). As we already saw, there is substantial evidence that
geography/geology, and history are clearly and closely related. Moreover,
geography, as Strabo sees it (again on the authority of Eratosthenes and
reaching back to Anaximander), is equally connected to both politics and
cosmology in a practical and theoretical context (1.1, 11). Now the Suidas
informs us that Anaximander wrote a treatise entited Tour of the Earth (G›V
p'r√odoV), and Athenaeus (11.498a–b) mentions a Heroology (≈Hrwolog√a).
As Heidel notes, Tour of the Earth was one of the first accepted names of a
geographical treatise,117 and Strabo’s references to Anaximander’s geographi-
cal work (and not just a “map”) on the authority of Eratosthenes certainly
appear to confirm this treatise (1.1, 11). Heroology, on the other hand, may
have been another title for Genealogies. Both were employed by (or attributed
to) Hecataeus, as Heidel correctly notes.118 Titles, of course, were not yet
employed at this time, so these references may have been part of his general
work on nature (i.e., chapters or sections in the general account), which
would include, in addition to cosmogonical and cosmological speculation, an
interest in the early history and geography of culture. After all, to explain (or
describe) how the present order of things was established (as it is clear from
Hesiod’s paradigm in the Theogony) entails offering an explantion of how the
present sociopolitical order originated.
This brings us back more specifically to Egypt. As we saw, one of the
primary differences between a mythical approach to the origin of humanity
and the rational approach is that the former assumes that humanity did not
have a real beginning in time but is the result of a series of events that took
place in illo tempore (or mythical time) involving supernatural entities, whereas
Gerard Naddaf 41
evidence in support of the Egyptian claim. Given the amount of silt depos-
ited in the Delta each year and the size of the Delta itself, the earth could
not be less than 20,000 years old, as Herodotus (2.13) and his Ionian prede-
cessors claimed, and thus the Egyptians themselves could very well have been
around for 341 generations or 11,340 years, as they claimed to Hecataeus
(Herodotus 2.143), with a statue of a high priest representing each genera-
tion.121 The difference between the two numbers could then account for the
period of “time” it would take/may have taken for the human species to adapt
to the land environment and discover the necessities of life before discovering
the various arts and crafts (I will discuss this in more detail later).
In the face of this, the Hellenic claim or conviction that humanity—or
at least the Greeks—originated sixteen generations ago looked, as Hecataeus
quickly realized, ridiculous. Hecataeus’ critical approach is reflected in the
opening of his Genealogies (G'n'alog√a): “Hecataeus the Milesian speaks
thus: I write [emphasis added] these things as they seem to me; for the stories
of the Greeks are many and absurd in my opinion” (FGH 1, frag. 1). The
spirit behind this critical statement is analogous to Xenophanes’ critical analysis
of the anthropomorphism that permeates the theology of Homer and Hesiod.
For his part, Hecataeus wants to rationalize the genealogies of the heroes, to
recreate a history of the past based on these genealogies and their respective
myths. There is no doubt that Hecataeus, like the Greeks, in general firmly
believed that the Homeric poems contained more than a kernel of truth. The
problem was separating the truth from the fiction, the rational from the
fantastic. The relatively new medium of writing would help to both record
the oral tradition and critically evaluate it. Indeed, Anaximander and Hecataeus
were among the first to have written their accounts in prose. Now, as we saw,
Hecataeus was already well aware of this because of Anaximander’s previous
research. Indeed, he must have read Anaximander’s prose treatise and observed
his map and the geographical treatise that accompanied it. Moreover, given
the dates, he may have even heard him lecture and/or narrate his logos, that
is, his rational and descriptive account. The question now becomes, to what
degree are Hecataeus’ Genealogies and Tour of the Earth based on Anaximander’s
Tour of the Earth and Genealogies/Heroologies?
The description of the earth and its inhabitants (oÎkoum°nh) is the
subject of the science called geography and, as W. A. Heidel judiciously
remarked, history and geography go hand in hand.122 This, moreover, is en-
tirely in keeping with our hypothesis according to which the logical point of
departure of a work of the p'r¥ f§s'wV type is none other than the society
in which man lives. In fact, it is clearly possible that Anaximander, the his-
torian, called the young in the Souda (˛ n'Ôt'roV ˘storik¬V),123 or simply
called the other Anaximander, the historian, by Diogenes Laertius,124 is none
other than Anaximander, the philosopher, of Miletus.125 This could be the
Gerard Naddaf 43
The legend of Danaus (king of Argos) and the Danaids (the sons or descen-
dants of Danaus) has a long history in connection to Greece and Egypt.
Martin Bernal (1987–1991) has argued in his much publicized series Black
Athena: The Afroasiatic Roots of Classical Civilization (vols. 1 and 2) that Greece
actually was invaded and colonized by Hyksos who were “thoroughly perme-
ated by Egyptian culture” (BA 2:45) in the late eighteenth century B.C.E. (BA
2:41, 363–64). Indeed, Bernal sees a similarity of the Greek hikes (ios), to
“Hyksos,” which, he claims, Aeschylus is punning in the play of that name,
Suppliants (≈Ik°tiV or ≈Ik°tid'V; see BA 1:22, 2:364). Bernal argues that both
Danaus and Cadmus (the two most prominent names for the case at hand)
were real Hyksos princes who settled in and created kingdoms in Argos and
Thebes, respectively (civilizing in the process its native inhabitants), and that
these kingdoms lasted until the end of the thirteenth century (2:47, 52, 58–
59, 502–504).127 He contends that this would tally with the report of Danaus
as the transmitter of the alphabet (2.502). It was during this period, he
believes, that much of the Greek language and culture was formed (2:525).
In fact, he contends that the Greeks themselves believed that they were
descended from the Egyptians and Phoenicians, and that they were, of course,
correct. Bernal is defending a diffusionist theory of culture, and this is precisely
what Anaximander appears to be defending, as we will see.
44 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
Meanwhile, the Danao√ or Danaans is, of course, one of the eponyms for
Greeks (Iliad 1.42, 56, 87, etc.), and Bernal is convinced that these are the
same as the Dene/Denyen (the Danauna), one of the infamous “Peoples of
the Sea” that, according to Egyptian documents, ravaged Egypt in the twelfth
century (BA 2.48, 59–60). It is unclear if the invasion of Egypt by the Danaans,
according to Bernal’s account, was perceived by the Danaans as a return to
the original homeland from which, according to fourth century B.C.E. histo-
rian Hecataeus of Abdera (whom Bernal cites in this context), their ancestor
and founder Danaus was expelled along with Cadmus as Hyksos leaders (BA
1:109; 2:503). Martin Nilsson, for his part, also sees a correspondence be-
tween the Danaans and the “People of the Sea,” and he believes that these
events were behind the myth of the Danaans.128 However, he conjectures that
the Danaids were the wives and daughters of these captured invaders whose
misfortune, as described by Aeschylus in the Suppliants, resulted from their
condition during captivity.129 According to Nilsson’s interpretation, the Danaans
(and thus the Greeks) are not direct descendants of the Egyptians, as Bernal
contends. Indeed, he clearly follows Aeschylus’ version in the Suppliants,
which Bernal finds “passionately chauvinist” with respect to the Egyptians
(BA 1.90). According to Aeschylus’ account, Zeus seduces Io (a priestess of
Hera) in Argos, but he is caught by Hera.130 Hera turns Io into a cow. Zeus,
however, continues to mate with Io by turning himself into a bull. Hera sends
Argos to guard the cow, but he is killed by Hermes. Hera then sends a gadfly
who drives Io all the way to Egypt. In Egypt, Zeus begets with her a son,
Epaphus. He in turn has a daughter, Libya, who has a son, Belos. He in turn
has two children, Danaus and Aigyptus, fathers, respectively, of fifty daugh-
ters and fifty sons. This is why the Danaids can trace their ancestory back to
Io and Zeus and thus to Argos and its present king, Pelasgos (who, moreover,
claims to be autochthonous: “I am the son of Palaichthon, child of the earth”
[Suppliants, 2420]).131 There is no doubt that, according to Aeschylus’ ac-
count, Egypt does not even benefit from its antiquity. Indeed, he clearly
demonstrates (or at least insists on) the priority of Argos and thus of Greek
culture and its civilizing role.132
But exactly when does Egypt enter into the picture? When did the story
become Egyptianized? Bernal insists that the proof “that the traditions of
Danaus and Cadmus go back to epic times” (BA 1:86) is found in a fragment
from the lost epic, the Danais, which describes the daughters of Danaus
arming themselves by the banks of the Nile, in order, one could assume, to
fight the sons of Aigyptus.133 However, this play, which probably was Aeschylus’
source since it also contains the Io episode, is generally dated to the sixth
century (or late seventh) B.C.e.134 Moreover, in the pseudo-Hesiodic Eoiai and
Catalogue of Women (c. 600), there is little, if anything, linking Egypt and
Greece, let alone Danaus and Aigyptus and/or Io. And in the Aigimios, Io’s
Gerard Naddaf 45
to Herodotus, the Phoenicians who came with Cadmus first introduced the
alphabet to Boeotia, Hesiod’s homeland (5.57.1–58.2). In fact, he contends, as
Bernal observes, that they also settled there. Herodotus is, however, ambiguous
about when this occurred. At 2.145, he mentions that the period of Cadmus’
grandson, Dionysus, goes back 1,600 years (…xak¬sia⁄t'a ka¥ c√lia) before
his time. This would entail that Cadmus and the Phoenicians introduced the
alphabet to Boeotia in the third millennium B.C.E., and thus prior to even
Bernal’s conjecture of around 1470 B.C.E.142 On the one hand, the approximate
period that Herodotus notes for the Trojan War in the same passage is close
to the current consensus: the thirteenth century B.C.E. On the other hand,
Herodotus contends that the Cadmean letters he saw at Thebes in Boeotia
were not that different from the Ionian. This would lead us to believe that he
was somewhat confused about how and when the transmission of the alphabet
occurred.143 Moreover, while most scholars associate the “mythical” (?) Cadmus
with Phoenicia, other traditions, as we saw, associate Cadmus with Egypt, as
in the Egyptianization of the story of Io.
Although Herodotus believed that the Phoenicians introduced writing or
grammata to Greece, he may have believed that the Phoenicians, in turn,
borrowed their writing system from the Egyptians (just as the Greek alpha-
bet, by adding vowels, represented an advance over the Phoenician writing
system). At 2.36, he clearly states that the Egyptians believed that their own
way of writing from right to left was superior to the Greek manner of writing
from left to right, and that they had both sacred and common or demotic
writing. There is no doubt, of course, that Herodotus believed in the diffu-
sion. Indeed, he believed that a considerable amount of Greek culture, includ-
ing their religion, was borrowed from the Egyptians (2.49–52). We could
legitimately assume that this also would be the case for the story of Io; that
is, Io is identified with Isis and was thus borrowed from Egypt.
On the other hand, Anaximander (and Hecataeus) argue that it was
actually before the time of Cadmus that the alphabet was introduced into
Greece, and the person who imported (m'takom√sai) it was Danaus (DK
12C1.11). There is no doubt here that Danaus is associated with Egypt and
its high culture. The fact that Hecataeus affirms that originally Greece was
populated by Barbarians led from Phrygia by Pelops and Egypt by Danaus
(FGH 1, frag. 119) lends credence to the idea that they thought that the
alphabet (or an alphabet) was introduced many generations before the date
that we now associate with its introduction. Anaximander may have traced
this in his own Heroology or Genealogies. But on what may Anaximander have
based his opinion on an Egyptian origin of the alphabet? Semiticists have no
problem calling West Semitic writing “the alphabet,” since each alphabetic
sign in the repertory stands for a single consonant and thus a phoneme, that
is, a class of sounds different enough from others sounds to change the
Gerard Naddaf 47
meaning of a word. But if West Semitic writing is an alphabet, can the same
be said about ancient Egyptian? According to Alan Gardiner, the Egyptians
very early developed a body of twenty-four uniconsonantal signs or letters
that he also calls “an alphabet.”144 In fact, he is convinced that this is the
origin of our own alphabet (1961b, 25–26). For the case at hand, it is not
important that some linguists and scholars may disagree with Gardiner. The
fact is, there is no good reason to believe that Anaximander was not con-
vinced that this was also the case after an Egyptian or someone else brought
this to his attention (how could he contest what he could not read?). The
Egyptians could demonstrate that writing had existed in Egypt even before
the Greeks could trace their first ancestors.145 Moreover, even if Anaximander
may not have been aware of the Linear B script, he may very well have been
aware of the Cypriote syllabary and thus some form of transition from the
introduction of an Egyptian alphabet to his own alphabet. In sum, this does
not exclude that Anaximander may still have thought that some individual
Greek genius innovated, at a more recent stage, by adding the five vowels to
the consonants—thus creating, as Powell notes, the first technology capable
of preserving, by mechanical means, a facsimilie of the human voice.146 There
is nothing to exclude that he saw Danaus and the Egyptians as the original
inventors of the alphabet, and yet the Greek alphabet as far superior to its
predecessors. The Greeks (or at least their intelligentsia) at this early stage
were already well aware of the powers of their own alphabet and sought its
true inventor. Thus Anaximander’s contemporary, Stesichorus (c. 630–555
B.C.E.), in the second book of his Oresteia, says that Palamedes invented the
alphabet ('•r–k'nai tΩ stoic'ƒa), that is, the Greek version of the alpha-
bet.147 Clearly, Anaximander and his generation see themselves as “writers”
heavily influenced, it is true, by oral tradition, but writers just the same.148
Herodotus, as noted earlier, believed that a great deal of Greek culture
and civilization originated in Egypt—a point also noted by Plato in the
Phaedrus (274c–d). And the famous statement by Hecataeus of Miletus, that
Greece originally was populated by barbarians brought from Phrygia by Pelops
and Egypt by Danaus, clearly means the same thing. This, moreover, is the
same Danaus whom he believed brought the alphabet with him. Now since
the opening remark in his Genealogies (FGH 1, frag. 1) strongly suggests that
he denied the gods any influence in civilization, then Danaus is seen as a
historical individual. Thus one function of his genealogies may have been to
retrace the origin of certain cultural icons with the help of information re-
ceived from Egyptian sources. In fact, if it is true, as Herodotus claims, that
Hecataeus attempted to trace his family back to a god in the sixteenth gen-
eration (2.141), and the Egyptian demonstrated to him that this was patently
absurd, then it was clearly the Egyptians who were instrumental in develop-
ing his critical approach and in giving him a clearer sense of chronology and
48 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
We saw earlier that the Nile and Egypt clearly had a special status for the
early Ionians. More important, given that they considered the Nile Delta, as
Gerard Naddaf 49
(or the Genealogies) entered into the picture? As we saw earlier, Anaximander
argued that an Egyptian cultural and/or political hero, Danaus, was respon-
sible for introducing the alphabet into Greece. Anaximander may have placed
a special importance on the alphabet, since he was well aware that his in-
quiry was founded on the testimonies that the alphabet allowed him to
collect.154 And we can conclude that this also was the case with a number of
other '•r–mata generally attributed to the Egyptians (e.g., stone monument
building). Somehow Anaximander acquired enough genealogical information,
presumably, but not exclusively, from Egyptian sources, to construct a sort of
chronological explanation of the diffusion of Egyptian culture.
Meanwhile, the section on the Tour of the Earth may have begun with the
Nile Delta, where life and civilization originated, and then proceeded either
clockwise or counter-clockwise about the orbis terrarum, the whole of Europe
and Asia and Libya, indicating the possible migrations of the various peoples
with which he was familiar from travels and various documented accounts.
The current location of each people also would have been sketched on the
map, as well as known routes for migration, trade, and military campaigning.
Of course, the question naturally arises about how he would account for
some peoples being clearly less civilized than others if Egyptian civilization
was transmitted with the migrations. The logical answer would be because of
the periodic occurrence of “natural catastrophies.” Indeed, even Thucydides
believed in the legendary flood, and there is evidence that Anaximander
believed that natural disasters were ongoing—albeit on a localized rather than
on a cosmic scale. Nor does this mean that there was nothing left to discover,
or that all past discoveries could be attributed to the Egyptians. Clearly,
Anaximander, one of the founders of the new enlightenment, was well aware
that his own rational approach was novel and exciting, indeed, far superior to
the current Egyptian approach. How could such a rationalist concur with
Egyptian religious practices without lamenting their deficiencies! As for the
fundamental differences between the different languages, Anaximander could
observe that given the fact that contemporary Greek dialects varied consid-
erably, it is not surprising that the Greek language appears considerably dif-
ferent from the Egyptian (or Phoenician). But given that certain similarities
could be attributed to both (as in the case of religious syncretisms), this would
be ample evidence that Greek language and civilization originated in Egypt.
Although we do not have a great deal of information on this aspect of
Anaximander’s historia, when the information is put into a historical perspec-
tive, it seems that Heidel and Cherniss (contra Guthrie [1962, 75] and the
vast majority of classical scholars) were not far off the mark when they con-
tended that the aim of Anaximander’s book was “to sketch the life-history of
the cosmos from the moment of its emergence from infinitude to the author’s
own time” (Heidel 1921, 287), or “to give a description of the inhabited earth,
Gerard Naddaf 51
geographical, ethnological and cultural, and the way in which it had come to
be what it is” (Cherniss 1951, 323). Let us remember again that what Hesiod
is attempting to do in the Theogony is explain how the present order of things
was established.
Meanwhile, an excellent example of this manner of proceeding comes to
us from Diodorus of Sicily, a historian from the first century B.C.E. What is
striking is that before undertaking his history of the Greeks (which will
include not only a chronological table of events from the Trojan War to his
own time but also events and legends previous to the Trojan War), Diodorus
begins with a cosmogony (1.7.1–3) and then moves to a zoogony (1.7.4–6),
and finally to a politogony (1.8–9). After briefly expounding these three
phases, Diodorus then turns to Egypt (to which he dedicates several books)
to start his history, strictly speaking, because tradition considers it the cradle
of the human species; his descriptions of the ideal conditions of the Nile are
analogous to what we saw in Herodotus (1.9ff.). It is worth noting that
Diodorus believes that numerous peoples had autochthonous origins, and
that this explains the origin of the diversity of languages (1.8, 3–4).155
What is certain regarding Diodorus’ account is that it is impossible to
attempt to attribute its contents to the influence of a “particular” philosopher.
In other words, the whole of the text is necessarily eclectic.156 However, the
text also is clearly of Ionian inspiration, and there is little, if anything, in it
that cannot be traced back to the sixth century B.C.E. and ultimately to
Anaximander. That one would think that Anaximander (as many scholars
contend) would not have reflected on the origin of language (albeit, clearly
on the origin of the alphabet), that he would have been incapable of initiat-
ing a theory of language as sophisticated as Democritus’, is, quite frankly,
astounding.
Diodorus, for his part, believes that history is the key to happiness
('¶daimon√a), since it commemorates the great deeds of past men and thus
incites us to emulate them, that is, to furnish us with examplars for noble
living. He thus contends that history is the prophetess of truth and the
mother of philosophy (1.2.1–2). The geographer, Strabo, also makes the same
contention when he asserts that philosophy and geography are both con-
cerned with the investigation of the art of life or happiness ('¶daimon√a,
1.1). This is the same Strabo who contends, on the authority of Eratosthenes,
that Anaximander was among the first geographers (1.1; 1.11). Meanwhile,
Strabo also begins with a cosmogony before introducing his text on Egypt
(17.1.36).157 Moreover, he claims, after mentioning Anaximander’s geographical
treatise (1.11), that the study of geography entails an encyclopedic knowl-
edge, and this includes a special knowledge of astronomy and geometry to
unite terrestrial and celestial phenomena (1.12–15). This connects with
Hipparchus of Nicaea’s (c. 150 B.C.E.) contention that it is impossible for any
52 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
beyond the Pillars of Hercules. And the distance from the Nile Delta to
where the Nile originated in the southern ocean would have to correspond to
the distance from the Nile Delta to the northern ocean.175 From this perspec-
tive, the distance from the Nile Delta to the Pillars of Hercules (and there
must have been a reasonable idea of this distance, given that the route was
often traveled) would have to correspond to the distance from the Nile Delta
to the southern ocean. These are some of the contraints that follow (or would
follow) if the map of the earth (inhabited or not) attributed to Anaximander
was indeed drawn as round and encircled by the ocean. Moreover, the size of
the seas also may have influenced Anaximander’s conception of the relative
sizes of the land masses of the three continents. The Nile and the Phasis
would be the natural divisions of the continents for Anaximander. Since India
and the Indus only appear to enter into the world picture with Hecataeus, the
Gerard Naddaf 55
see Gantz (1993, 164–65); the tradition goes back at least to Hesiod; see frag. 2, 4,
6, and 234 (ed. Merkelbach and West).
12. For other versions and a discussion, see Gantz (1993, 165ff.). The version
here contains analogies with the flood that destroyed Atlantis (see Naddaf 1994).
13. See Pindar, Olympian 9.42–53. He seems to be echoing Acusilaus (c. 500);
see Gantz (1993, 165).
14. G. S. Kirk (1974, 136), for his part, sees in this an etymological creation:
løaV (stone), la¬V (people). This, it seems to us, amounts to the same thing.
15. It seems that this myth was adopted by most Greek cities to explain the
common origin of mankind. As Apollonius Rhodius notes, Deucalion was the first to
found cities, build temples, and rule over men (Argonautica 3.1085–86; for another
interpretation, see Nicole Loraux (1991, 392).
16. See, for example Hippolytus, 12A11; Alexander, 12A27; Aetius, 12A27, 30;
Aristotle, 12A27; Censorinus, 12A30.
17. Aetius, 12A30. The language and the terminology (floioƒV p'ri'c¬m'na,
p'rirrhgnum°nou to£ floio£) suggest that Anaximander saw an analogy between
the development of the animal and that of the universe—or vice versa.
18. For another reading of ÷p ’ ıl√gon cr¬non m'taiÍnai, see Guthrie (1962,
102). According to his interpretation, animals “lived on for a short time.” I fail to see how
this makes any sense. For an excellent discussion of this passage, see Conche (1991).
19. See Hippolytus, Refutations 1.6.6 = DK 12A11.
20. Conche (1991, 222).
21. Barnes (1982, 22); see also DK 12A27 = Alexander of Aphrodisias; and
Aristotle, Meteorology 353b5.
22. DK 12A10, 37–40.
23. DK 12A11, 16–17.
24. DK 12A30, 34–37.
25. DK 12A30.
26. See the quote from Pliny and Lucretius in Conche (1991, 229–30). For an
interesting account that contrasts in a way with Anaximander’s, see Lucretius, On the
Nature of Things 5.222–25; 800ff.
27. The passage may, however, be translated as “originally, humans were born
(g'gon°nai) like another animal, namely a fish.”
28. For an analogy, see Diodorus’ description and Kahn’s (1960, 112–13, 70–71)
comment.
29. Plutarch alludes to the galoi or squales, of which one species, called the
“smooth shark,” has the remarkable particularity of the fetus being attached to the
mother’s stomach by an umbilical cord. This is an organ analogous to the placenta; see
Table Talk 730e.
58 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
30. The species in question is the subject of one of Aristotle’s most famous
descriptions in History of Animals 565b1. He notes that the young develop with a
navel string attached to the womb.
31. According to Mansfeld (1999, 23), Plutarch and Censorius may only be
commenting on Aetius’s summary in his Placita.
32. Kahn (1960, 112–13, 70–71), for his part, does not share the idea that
Anaximander believed man to be born from or to come from another species.
Anaximander could very well assign the original human embryo to floating mem-
branes, such as we find them in Diodorus of Sicily 1.7. In effect, according to Kahn,
Anaximander must have believed that the origin of man was entirely comparable to
that of the other terrestrial animals. In this regard, Pseudo-Plutarch’s sentence, “man
is engendered by animals of another species,” does not mean, following Kahn, that
these living beings of “another species” do not distinguish themselves from the original
form of the other animals. It simply means that they are different from human beings,
such as we know them.
33. See Adkins (1985, 59); Burkert (1985, 248).
34. Iliad 1.97, 1.220–22, 1.531.
35. For the elders in the boule, Iliad 19.303; for the full council in the agora, Iliad
18.497ff.
36. Ònax, Iliad 3.351, or the lord of counsel or all wise! (mht√'ta, Iliad 1.175;
Hesiod, father of gods and men (pat‹r ™ndrÍn t' q'Ín t', Theogony 542). Victory
always entails violence and power. The gods battle one another at Iliad 20.31ff.; they
are always angry with one another when one god gives the advantage to another. Hera
is a prime example; tricking Zeus and Zeus’ reaction, Iliad 15.1ff.; feasting, Iliad 19.165.
37. Ònax ™ndrÍn, Iliad 1.506; 9.114.
38. Theogony 391ff.
39. Could this new conception of justice be applied without a written code of
laws? Havelock (1978, 19) sees this newly emerging principle of justice and order in
the universe as being due to the passage from an oral to a written culture. According
to him, an oral culture is incapable of conceptualizing justice outside of a pragmatic
application of daily procedure, so the justice of the nobles is not the justice of Zeus,
although they themselves may see it as justice, that is, as part of their own desserts,
as a question of tim– or honor (d√kh is what the nobles have the right to expect
between given persons in given situations). Murray (1993, 61), for his part, contends
that Hesiod also created a political vocabulary.
40. Vernant (1983, 206–208); see also Detienne (1996, 89–106).
41. While the first inscriptions were of a strictly “private” nature (see, e.g., Robb
[1994, 44–45, 274–75]), with the advent of written law, writing becomes very much
a “public” affair (see Detienne [1988, 41]) and thus may well have encouraged mini-
mal literacy among the “people.”
42. For a good discussion, see Willetts (1977, 216–23), and, more recently, Robb
(1994, 99–124). One interesting feature about the advent of law codes, as Marcel
Gerard Naddaf 59
Detienne has noted, is that they always were exposed in a “public space” for all to see
(1988, 41). In sum, seeing was more important than reading.
43. According to Coldstream (1977, 302), this is the earliest known example of
“alphabetic writing being pressed into the service of the polis.” Gagarin (1986, 86)
contends that, “the main purpose of the Drerian law is to prevent the judicial process
from being corrupted or otherwise abused for political or financial gain.” I largely
concur with this. See also Robb (1994, 84ff.). For a very different position, see Osborne
(1996, 186), for whom the primary purpose of the law (indeed, written law codes in
general) was “to control the distribution of powers within the elite.” It is thus an elite
“self-regulation.”
44. Jeffrey (1990, 90); on the Euboean role in the introduction of the alphabet,
see Powell (1997, 22, and note 39).
45. The fact that the laws of Charondas and Zaleucus had a great deal to do with
regulating commerical transactions is testimony to the importance of trade (see Gagarin
1986, 65–66).
46. But certainly defense, trade, and land still were factors in the settlement of
any site. Thus Corinth’s capture of Corcyra from the Eretrians in 733 on their way
to founding Syracuse was probably a defensive move.
47. Since the first inhabitants did not bring women with them, they must have
later married with local women and/or eventually brought women from their former
homeland. Ironically, in the new colonies, since land, strictly speaking, was no longer
a problem, he who had the most sons could increase his family wealth the fastest.
48. Van Wees has argued that in Homer (see, e.g., Iliad 2.198–202), there is
already “an embyonic hoplite phalanx in the process of developing” (1997, 691).
49. There is one passage in the Theogony where Hesiod does employ the term
techne (861–64). This rarely quoted passage is interesting for the present purpose, since
it describes young men in a foundry learning the art of metallurgy.
50. Indeed, Herodotus notes (2.167) that in his own day the Corinthians had less
prejudice than any other Greeks against craftsmen.
51. The famous Chigi vase (c. 650), which originated in Corinth during the reign
of Cypselus (c. 657–27), would seem to confirm this. It depicts a clash of hoplites. For
an interesting discussion on the vase, see Osborne (1996, 161–64).
52. On the dating of Cypselus and Periander, see Grant (1988, 84–85). See also
Dillon and Garland (1994, 35).
53. On Solon and the act of writing, see Naddaf (2000, 348–49).
54. I am beginning here with the tradition of the Ionian foundation of Miletus
during the famous resettlement period that followed the collapse of the Mycenaean
world. I do not mean to imply that Miletus did not have a Bronze Age past. The
archaeological evidence clearly shows that the site was almost constantly occupied
from at least 1700 B.C.E. There also is a later tradition on the foundation of Miletus
by second-century C.E. traveler, Pausanias (7.2.5). According to his version, the city
was founded by a Cretean named Miletus. He landed there with his army while
60 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
fleeing Minos, the son of Europa. For a detailed discussion of the Bronze Age period
and other versions of the foundation of Miletus, see Gormon (2001, 13–31).
55. Scholars have not reached a consensus on the date of the Lelantine war. For
a recent discussion, see Parker (1997, 59–83). In his Works and Days (654–59), Hesiod
boasts of achieving a poetic victory in the funeral games at Chalcis in honor of King
Amphidamas. Plutarch (Moralia 153 e–f ) informs us that Amphidamas lost his life
during the war.
56. On the aristocratic nature of the Lelantine war, see Jeffrey (1977, 209),
Murray (1993, 76–79) and, most recently, Thomas and Conant (1999, 188–89).
57. For the evidence, see Konon FGH 26 F 44 and Nicolaus of Damascus FGH
90 F 52.
58. Herodotus (5.92) certainly describes the Bacchaids as an oligarchy.
59. For references and a discussion, see Dillon and Garland (1994, 11–12).
60. Murray (1993, 148) contests the use of the term polemarchos as military leader
at this point in time; see also McGlew (1993, 71–72); not everyone agrees (see e.g.,
Jeffrey 1977, 147).
61. Herodotus (5.92b2) states that the Delphic oracle suggested that Cypselus
would bring justice to Corinth (dikaiÔs'i dÆ K¬rinqon). As Forrest (1986, 22)
notes, the word dikaiÔs'i entails more than a hint here of “equality.”
62. Herodotus says that Ardys ruled Lydia for forty-nine years (1.16), but this
seems far too long when considering the dates of his successors Sadyattes, Alyattes,
and Croesus. Huxley (1966, 75) suggests about thirty-five years for the reign of Ardys
(therefore, c. 652–614); Grant (1988, 290) proposes twenty-seven years (therefore, c.
652–625). Dating generally is confusing. Since Herodotus strongly suggests that the
Lydian king, Aylattes, and the tyrants, Thrasybulus of Miletus and Periander of Corinth,
were very much involved with one another (see below) at the beginning of the reign
of Alyattes, and since there is a good deal of consenus that Periander’s tyranny runs
from about 628 to 586, I would suggest the following: Ardys (c. 652–629), Sadyattes
(629–617), Alyattes (617–560), and Croseus (560–546). For Alyattes and Croesus,
compare Herodotus 1.25 with 1.86. Herodotus says that the reign of Sadyattes lasted
twelve years (1.17).
63. Since it was the countryside that was ravaged, it meant that the landowers
had the most to lose, therefore, one would expect that a deeper sense of community
was maintained.
64. According to Herodotus (1.20–22), it was actually Periander who helped
Thrasybulus trick Alyattes into making peace with Miletus.
65. As Salmon notes in his article “Lopping Off the Heads,” the famous story
of Thrasybulus (or Periander) taking the messenger into the field and silently cutting
off the tallest ears of corn and throwing them away may be simply a metaphor for the
establishment of equality: lopping off the heads of the so-called preeminent citizens
may assure equality among the remaining citizens (1997, 60).
Gerard Naddaf 61
66. See Hahn in this book; also see Jeffrey (1977, 211).
67. Jeffrey (1977, 214) believes that he may have lived until the middle of the
sixth century. Of course, in this case, she sees the Lydian invasions of Milesian ter-
ritory as having occurred around 600 B.C.E. Since it is clear that Thrasybulus had a
xenia or guest-friendship pact with Periander (which may have been Miletus’ saving
grace in the war with Alayattes), this may have influenced future accounts about their
relationship, including the one about how to maintain power. In fact, if Periander
became more ruthless toward the end of his tyranny, this may have influenced the
accounts about Thrasybulus because of their xenia. Thrasybulus already may have
disappeared by this time.
68. Plutarch, Greek Questions 32 = Moralia 298c and Herodotus 5.29, who says
sixty years. This is the only reference in the extant sources of Thoas and Damasenor.
For an excellent and a detailed discussion of this passage, see Gorman (2001, 107–21).
Instead of reading Tischner’s emendation of turannos from the genitive plural to
accusative plural, Gorman retains the accusative and reads Plutarch as referring to “the
tyrants around Thoas and Damasenor” (tÍn p'r¥ Q¬anta ka¥ Damas–nora
turºnnwn). She then argues that the word “tyrants” is employed in the sense of an
oligarchy such as the famous Thirty Tyrants at Athens. Gormon situates this oligarchy
in the eighth and seventh centuries B.C.E., or directly after the Neleid monarchy had
run its course. Consequently, she places the famous Milesian stasis and subsequent
Parian mediation in the seventh century, that is, before the tyranny of Thrasybulus
(120). However, it is difficult to reconcile this thesis with the fact that the Neleids
were still clearly in power during the Lelantine war (c. 700) and/or the fact that Gyges
(not to mention his successors) tried to overtake Miletus around 680 B.C.E., and failed.
I will develop another thesis later.
69. For an interpretation, see Murray (1993, 248); Grant (1988, 159–60); Hurwit
(1985, 205); Huxley (1966, 79–80); Jeffrey (1977, 214); Gorman (2001, 108–10). Jeffrey
contends that the fighting was between two aristocratic clubs (…tair√ai). Athenaeus
(12.523f–524b), citing Heracleides of Pontos, gives a brief description of the more
gruesome details of the stasis between the rich and the poor. However, Gorman (2001,
102–107) has convincingly shown that this description lacks credibility.
70. On this, see Starr (1977, 148–52). From this perspective, the members of the
new oligarchy would not necessarily be farmers, strictly speaking. If the oligarchy were
restricted to farmers, the class of merchants, artisans, mariners, and so on would have
been deprived of political power. Given the period and the wealth created by this
group, this seems unreasonable.
71. Jeffrey (1977, 214) sees the tensions as actually heightening after the fall of
Croesus and thus places the Parian resolution as later. This shows to what degree there
is a lack of consensus on Milesian history.
72. According to Murray (1988, 461–90), the Ionians revolted from the Persians
because they were making them poorer. However, Georges (2000, 1–39) recently has
shown that the Ionians in general and the Milesians in particular actually prospered
(or rather continued to prosper) under the Persians. He argues that the revolt was
62 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
connected to the fact that the political classes of Ionia were alienated and paralyzed.
The tyrants lost their former close connection with the people. What led the tyrant
Aristagoras of Miletus to revolt was the fact that the Persians eventually began to treat
the tyrants themselves as dispensible vassals rather than as equals.
73. This, in turn, would initiate an even greater knowledge among the Ionians
of Asia, culturally and geographically. Indeed, Cyrus employed Ionian masons, sculp-
tors, and architects to work in his sumptuous new palace in Persepolis.
74. This actually occurred for a brief period between 500 and 494. The Milesians
certainly resisted the return of Histiaeus around 500 B.C.E. The experience with “de-
mocracy” ended after the sack of Miletus by the Persians in 494 B.C.E. See Huxley
(1966, 149–50).
75. This also is suggested by Jeffrey (1977, 219). She hints that Histiaeus may
have been among the group chosen by the Parians. This makes perfectly good sense
although, in my view, her dates for the conflict between the two groups (aristocratic
in her eyes) are too late.
76. Phocylides is clearly against forms of extreme insatiable wealth. This also is
expressed in frag. 5, where he contends that a small, well-ordered city on a rock is
superior to foolish Ninevah. Phocylides is clearly aware of the once mighty Nineveh
that was destroyed in 612 (West 1978, 166; Snodgrass 1980, 175). Herodotus makes
a similar comment with respect to the size of Babylon. It was so large that the enemy
entered its gates without those in the city center even being aware (1.191).
77. On the league, see Jeffrey (1977, 208–209).
78. West (1978, 164–67) makes a number of good arguments for dating Phocylides
around the same time as Thales and Anaximander. For some reason, Gormon (2001,
73) situates Phocylides before 650.
79. Agathemerus, Geography 1.1 = DK 12A6, 68B15.
80. Heidel (1937, 132; 1921, 247) argues that the sense of the phrase implies that
Anaximander’s map was accompanied with a written treatise. Kirk (1983, 104) and
Conche (1991, 25, n. 3) contest this. In my view, the ancient testimony clearly implies
that the map was accompanied with a written treatise explaining its function.
81. Strabo, Geography 1.1.11 = DK 12A6, 30–34. On Anaximander, also see
Strabo’s opening remarks at 1.1.1: “Geography, which I have now chosen to con-
sider, I hold as much the pursuit of the philosopher as any other science. That my
opinion is sound is clear from many considerations. For not only were the first
who boldly essayed the subject men of this sort—Homer, Anaximander, and
Hecataeus (as Eratosthenes also says) . . . this task belongs peculiarly to the man
who contemplates all things divine and human, the science of which we call
philosophy.”
82. On this, see Christian Jacob, “Inscrire la terre habité sur une tablette,” in
Detienne (1988, 276–77). Jacob argues that the map was only a theoretical object, and
that its construction was not based on empirical data (281). This is a rather extreme
position.
Gerard Naddaf 63
83. On Divination 1.50.112 = DK 12A5a. Cicero does not say how Anaximander
did this, although he explicitly says that it was not an act of divination (indeed, he
calls him the “physicus”). Anaximander could have based his prediction on the obser-
vation of abnormal animal behavior.
84. Universal History; fr. 27 = FGH 3 frag. 581; see Diogenes Laertius 2.2; see
also Eusebius 14.11 = DK 12A4, who adds “times and seasons” to the solstices and
equinoxes. I discuss this in more detail later.
85. Although Croesus was successful in convincing the Spartans, it was not for
the same reason given by Herodotus. According to Herodotus, the Spartans agreed to
help Croesus because of a gift of gold that he had given to Sparta a generation earlier
to use for a statue for Apollo (1.69).
86. See, for example, Jucker (1977, 195–96); Gelzer (1979, 170–76); Yalouris
(1980, 85–89); Hurwit (1985, 207–208); Conche (1991, 38–41).
87. At least it is strongly suggested by such epithets as ÒyorroV or “backward-
flowing” to qualify Oceanus (e.g., 18.399; see also Iliad 14.200–201). This also is
derived from the description of Achilles’ shield (18.607) and was clearly Strabo’s
interpretation (1.1.3). Indeed, he refers to Homer as if he constituted an important
authority on the matter (1.1.7).
88. On the rationalism, see Lévêque and Vidal-Naquet (1997, 52–55).
89. See DK 12A 11, 23, 24. The impact of Anaximander’s rational/natural ap-
proach to meteorological phenomena is best described by Aristophanes in Clouds 404ff.
90. There is no reason to think that Anaximander would not have been aware of
the same story that Herodotus relates (4.42) about the pharoah, Neco (609–594),
sending several Phonecian ships to circumnavigate Africa/Libya.
91. See, for example, Lévêque and Vidal-Naquet (1997, 80), and, more recently,
Couprie in this book.
92. As Conche notes (1991, 46, n. 47, 48). According to Georges (2000, 11),
“Didyma was to archaic Ionia what Delphi was to the Greeks of Europe.”
93. However, at 4.110, Herodotus employs oÎkoum°nh and not g› to refer to the
inhabited earth.
94. The Nile, Phasis, and Tanais are explicitly mentioned by Herodotus at 4.45;
for the Tanais, see also Hecataeus (FGH1, frag. 164, 165). Pindar also mentions the
Phasis and the Nile as the northernmost and southernmost borders of the Greek
world, at Isthmia 2.41ff. The Phasis would be seen as flowing from the Caspian Sea
which, in turn, was seen as the bay of the ocean. The Ister or Danube also has a strong
claim according to Heidel (1937, 31–44); see also Herodotus (5.9; 4.46–50). Accord-
ing to Heidel (1937, 21), the Ister and Nile correspond to the tropics.
95. As I noted in note 90, there is no reason to think that Anaximander would
not have been aware of the same story that Herodotus relates (4.42) about the pharoah,
Neco (609–594), sending several Phoenician ships to circumnavigate Africa/Libya.
This is a more “secular” belief that the continents were surrounded by water.
64 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
96. In reality, Herodotus states that this is what the “Ionians” believe, but it
seems from the context that the Egyptians also thought that the Nile Delta was where
they originated.
97. This contention is clearly expressed by Diodorus of Sicily (1.10.1). See also
Plato, Timaeus 22a–23a, Aristotle, Meteorology 1.14.352b19–21; and Politics,
5.10.1329a38–b35. On the North of Egypt as having the “ideal” climatic conditions,
see Hippocrates, Aphorisms 3.1. Thales’ famous contention that life arose from the
primordial waters may be of Egyptian origin. Indeed, there is no reason to distrust the
doxographical claims that he visited Egypt for reasons discussed below. Aeschylus,
meanwhile, seems to be the exception, despite references in the Suppliants clearly
reflecting the Nile’s fertility in the context of human origin: “Native stock spawn from
the fertile Nile” (270); “the race the Nile breeds” (485).
98. One possible allusion to a fabulous past is at Iliad 9.181–185, where Thebes
is described as a city of 100 gates and where treasures are in greatest store.
99. On this point, see Froidefond (1971, 64–67). According to Froidefond, the
only references to Egyptian history in the Odyssey would date to the last part of the
eighth century. Heubecks/West/Hainsworth note in their commentary on Homer’s
Odyssey that the choice of the name Aigyptus for a minor figure at Odyssey 2.15
“reflects the novelty of reopened communications with Egypt” (1988, 130 and see also
192). Morris, for her part, argues that a number of episodes reflect the end of the
Bronze Age, “when foreigners of many lands attacked the Egyptian Delta and en-
gaged the Pharaonic forces throughout the Ramesside period” (1997, 614).
100. For a good recent summary of the Greeks in Egypt, see Boardman (1999,
111–59).
101. Gardiner (1961a, 362), notes that in order to appease the native Egyptians,
Amasis restricted the merchant activity of the Greeks to Naucratis. On the other
hand, as Herodotus notes (1.29), both Croesus and Amasis attempted to attract to
their respective courts the intellectual elite. Given the period, Solon, Anaximander,
and many others may have met there.
102. The Saite dynasty included Psammetichus I, Necho (610–595), Psammetichus
II (595–89), Apries (589–70), and Amasis (570–26).
103. See Froidefond (1971, 71); Boardman (1999, 141).
104. Boardman (1999, 132). Naucratis fell to the Persians around 525, and the
archaeological evidence indicates that after the Persian invasion of Egypt, the relation
between Greece and Naucratis, at least until round 500 B.C.E. was severely affected
(141). However, there does not seem to have been a decline in the economic fortunes
of the Ionian states in general and Miletus in particular under the Persian rule of
Darius, that is, until the Ionian revolt of 500 and its aftermath. It also is worth noting
that Darius may have lifted the pharaonic travel restrictions, thus making the rest of
Egypt more accessible after 525. On the other hand, given that Miletus maintained
a privileged relation with the Persians, as we saw earlier, it is unclear why Darius
would have supressed commercial relations with Naucratis, unless Naucratis itself
strongly supported the Egyptians during the conflict.
Gerard Naddaf 65
105. Hurwit (1985, 184); see also Boardman (1999, 143–44). There is evidence,
as I noted earlier, that Thrasybulus already may have initiated similar temple building.
106. According to tradition, Deucalion lived only a few generations before the
Trojan War; see Gantz 164ff.; also see Thucydides above.
107. This certainly was Plato’s conviction, and there is no good reason to believe
that this idea did not have a long history (see Naddaf 1994, 192–195; Naddaf and
Brisson 1998, xxiii, xxvi–xxvii).
108. Arrian, Anabasis 5.6.5 = FGH 1, frag. 301; Strabo 12.2.4; see also Heidel
(1943, 264).
109. For a comparision with Anaximander, see Kirk (1983, 140).
110. Guthrie (1.362–63); Kirk (1983, 163–64); Lesher (1992, 1).
111. Xenophanes recalls the coming of the Mede in fragment 22: “How old were
you when the Mede arrived?” (DK 21 B22). The fact that Apollodorus states that
Anaximander was sixty-four in 547–46 (= Diogenes Laertius 2.2) suggests that he
may have been using the fall of Lydia and/or Ionia to the Mede as a reference. Indeed,
Anaximander may have made a reference to this in his book.
112. Plutarch suggests he did; see Isis 379b and On Superstition 171d–e.
Heidel (1943, 274) notes that the stories of Xenophanes visiting Egypt are prob-
ably apocryphal.
113. Naucratis attracted poets such as Sappho and Alcaeus, artists such as Rhoikos,
statesmen such as Solon, philosophers such as Thales, and so on in addition to get-
rich-quick traders. See Boardman (1999, 133).
114. Heidel (1943, 262). In Heidel (1921, 243), he states that, “Hecataeus was
only a trifle over a generation younger than his fellow townsman Anaximander,”
although the rationale for this on page 260 gives a similar date of 560 for his birth;
on the other hand, in Heidel (1935, 120), he states that Hecataeus probably was born
shortly after Anaximander’ death. There seems to be a problem. Meanwhile, Hecateaus’
dates are similar to Heraclitus’ (555–480). If Anaximenes (580–510) was a pupil of
Anaximander’s (see Diogenes Laertius Lives 2.3), then this also would suggest
Anaximander’s wide range of interests.
115. Heidel (1943, 263) believes that Hecataeus visited Egypt with Cambyses on
his expedition of conquest.
116. See Heidel (1943, 262). We must remember that the phusikos Anaximenes
also is said to have been his pupil.
117. Heidel (1921, 241).
118. Ibid., 262.
119. The Egyptians contend, according to Diodorus, that this continues to occur
for some forms of animal life (see Diodorus 1.10). There is an astonishing similarity
between his account in 1.10 and Herodotus’ account in 2.13, which may explain the
correlation he drew between humans and fish. For an interesting discussion on the
reproduction of fish in Egypt, see Herodotus 2.92–94.
66 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
120. For a discussion of the importance of chronology for the Egyptians, see
Gardiner (1961b, 61–68). He also discusses the famous Palermo Stone.
121. The geological speculation may have been initiated by certain phenomena
taking place around Miletus itself (see Kirk 1983, 139). Egyptian phenomena, on the
other hand, provided a way of quantifying the hypothesis.
122. Heidel (1921, 257).
123. DK 58C6 = FGH 1, frag. 159.
124. DK 58C6, 23–24 = Diogenes Laertius, Lives 2.2; see also DK 12A1.
125. This also is Delattre’s opinion in Les Présocratiques (1988, 589, n. 2).
126. The doxography is taken from a scholium on the second-century B.C.E.
grammarian, Dionysius Thrax (= p. 183.1, ed. Hilgard).
127. According to Bernal’s reconstruction, the Hyksos invasion was the second
time that Greece had been colonized by Egyptians. The first time was during the third
millennium, when the Egyptian colonizers brought their advanced building tech-
niques, cults, and religion (see, in particular, 1987, chaps. 2 and 3).
128. Nilsson (1932, 64ff.). A number of scholars argue that there is no relation
between Homer’s Danaoi and the Dene/Denyen or “Sea People.” See, for example,
Tritle (1996, 325).
129. Thus according to Nilsson, to regain their freedom, they may have massa-
cred the Egyptians, who had held them as concubines, and this would explain the
unsavory role played by Aigyptus and his sons.
130. On the genealogy of Io, see Gantz (1993, 198–204).
131. Although the Danaids insist that their origin is Argive (Suppliants 323),
Danaus himself is well aware that he/they are physically different from the Greeks
(496–98, 270ff.). However, they also are aware of being of one stock, the products of
the Nile (281ff.).
132. Thucycides’ position is that the sons of Hellen hellenized the indigenous
Pelasgians (1.3), which again is inexcusable for Bernal (1987, 101–103).
133. Frag. 1, Poetae Epici Graeci 1 (ed. Bernabé).
134. Hall (1996, 338); see also Séveryns (1926, 119–30).
135. West (1985, 145–46, 150). See also Hall (1996, 338).
136. Timothy Gantz’s Early Greek Myth: A Guide to Literary and Artistic Sources
(1993) is indicative of the competing and conflictive claims. This, of course, is not
unique to ancient Greece but appears to be the norm in most ancient cultures.
137. For the genealogical intricacies concerning Io and a reasonable discussion,
see Gantz (1993, 198–204).
138. Lloyd (1975–88, 1:125); Hall (1996, 338).
139. Hall (1996, 338).
Gerard Naddaf 67
140. For my own critique of Bernal’s position and Afrocentrism in general, see
Naddaf (1998b, 451–70).
141. See, for example, Snodgrass (1971, 351); Coldstream (1977, 342ff.); Powell
(1997, 18–20); Burkert (1992, 25–26).
142. For some reason, this is not the date that Bernal (1991, 361) gives for the
arrival of the Phoenician or Cadmean settlement in Thebes, but around 1470 B.C.E.
143. See Coleman (1996, 286), and Tritle (1996, 326), who notes that it is
somewhat surprising and suspicious.
144. Gardiner (1961b, 23).
145. Perhaps this is what Plato has in mind in the Phaedrus (274e) when he
attributes the invention of grammata to the Egyptian Theuth—a story, moreover,
which takes place near Naucratis (274c).
146. Powell (1997, 25).
147. Frag. 213, Poetae Melici Graeci (ed. Page).
148. When Aeschylus states in his ode to progress in Prometheus Bound (460–61)
that writing (grammata) is the memory of all things and thus the productive mother
of the arts, the alphabet is clearly not perceived as a relatively new invention that in
reality it was.
149. Anaximander, for his part, would have insisted on the fact that there would
be no room for gods in his genealogical analysis.
150. For a recent discussion on this controversial fragment, see Lesher (1992,
149–55).
151. See Huxley (1966, 20); Colophon also has an early Egyptian connection
through the travels of Mopsus (20).
152. See Huxley (1966, 20).
153. The popular theme of the fecundity of the Nile is found in Aeschylus’
description of it as fus√zooV (Suppliants 584).
154. See Thomas (1992, 114) on kl°oV (glory) and writing. Where would
Anaximander fit in?
155. On the other hand, he notes at 1.9.5 that the fourth-century historian,
Ephorus (already mentioned earlier in the context of the alphabet), believed that the
barbarians were prior to the Greeks, which suggests that he held a position similar to
the one that I have attributed to Anaximander.
156. The origins of Diodorus’ text have provoked a vivid controversy. For a
discussion, see Burton (1972).
157. The text echoes back to 1.3.4 and also is concerned with the phenomena of
shells that one observes in certain regions of the country.
158. This also is suggested by Thompson (1948, 97–98). Couprie, for his part,
appears convinced that Hahn and Heidel are correct, and he attempts a reconstruction
68 ANTHROPOGONY AND POLITOGONY IN ANAXIMANDER OF MILETUS
of such a map in this book (Fig. 3.16). I attempt a similar rendition later, but using
the Nile Delta as the center.
159. Parker (1974, 67) and Hahn (2001, 207) state that the Egyptians were the
first to divide the day and night into twelve hours each, and that they had a high
degree of astronomical knowledge, but this is highly contested by Neugebauer (1975,
2:560).
160. Heidel (1921, 244) or something analogous to what Thales did with Egyp-
tian geometry.
161. As Lloyd notes (1991, 293), the length of the solar year was only deter-
mined fairly accurately by Meton and Euctemon in about 430 B.C.
162. Robinson (1968, 32); Hurwit (1985, 208); Couprie, Figure 3.16 in this
book. Although Hurwit and Robinson explicitly state that the land was divided into
three equal parts, with the Nile and Phasis rivers as the dividers, in their reconstruc-
tion, Europe appears clearly larger than Asia and Asia clearly larger than Libya. This
is a perfect example of how difficult it is to reconstruct three equal land masses when
using the Nile and Phasis as dividers.
163. Brumbaugh (1964, 22); Conche (1991, 47, Fig. 2); Thompson (1948, 98,
Fig. 11).
164. Robinson (1968, 32); Hurwit (1985, 208).
165. Brumbaugh (1964, 22); Vidal-Naquet et Lévêque (1996, 53 ); Couprie, (in
this book).
166. Conche (1991, 46); Froidefond (1971, 167).
167. Conche (1991, 47, Fig. 2); Thompson (1948, 98, Fig. 11).
168. Conche (1991, 46); Heidel (1937, 11–12).
169. Brumbaugh (1964, 22); Robinson (1968, 32); Conche (1991, 47); Hurwit
(1985, 208).
170. Myres (1953, 6 Fig. 5); Heidel (1937, 1, Fig. 11).
171. J. O.Thompson (1948, 97, Fig. 10).
172. Couprie (in this book) gives a lucid explanation of what Anaximander could
achieve with the added and important reminder that for Anaximander the earth is flat
and not spherical.
173. Herodotus informs us (2.32) that it is a four-month march from Elephan-
tine to the Deserters. The Deserters is thought to be Sennar, which is around 150
miles south of Khartoum. Elephantine is approximately midway between the Delta
and Khartoum. At this point, Herodotus says that the river changes course, and no
one has gone beyond this point because of the heat.
174. At 2.33, Herodotus mentions the Danube/Ister flowing through the center
of Europe (to the Milesian colony of Istria on the Black Sea) and being equal in
length to the Nile. More important, he states that Egypt, and thus the Nile Delta, is
more or less in line with the Cilician Mountains, Sinope, and the mouth of the
Gerard Naddaf 69
Danube/Ister. At 2.26, he suggests that the Danube/Ister and the Nile take rise on the
same degree of longitude. For a discussion, see Heidel (1937, 24–25); Thompson
(1948, 98).
175. Of course, no one had ever traveled to the end of the Nile. But there is no
reason to believe that Anaximander was not aware of the famous story about the
circumnavigation of Libya (Herodotus 4.42). Although Thompson (1948, 72) doubts
this story, he does believe that the time it is said to have taken (around three years)
would correspond more or less to the reality. If this were the case, given the param-
eters of Anaximander’s map, then there would be no reason to conclude that this was
evidence that Libya was far smaller than Europe or Asia, as Herodotus appears to
contend, according to Heidel (1937, 28). Clearly, Herodotus has a very different
conception of the size of Europe from his predecessors. Meanwhile, if the circumnavi-
gation did occur, one would expect that the Phoenicians also would have reported that
at one point during their southern voyage the temperature actually got cooler.
176. Miletus is approximately midway between the mouth of the Nile and the
mouth of the Danube/Ister. Of course, Miletus is not on this meridian, but it is a lot
closer than Sinope (see note 174) and, given the time frame, Anaximander may have
thought that it was.
This page intentionally left blank.
Proportions and Numbers
in Anaximander and
Early Greek Thought1
Robert Hahn
PROLOGUE
73
74 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
the influence of the cult of Amun at Napata, saw themselves as the legitimate
successors of the pharaohs. The appropriation of ancient forms acknowledged
a traditional role in the history of Egyptian authority. Consequently, during
their sixty-three years of reign, Kushite rulers reinvigorated Egypt in those
established traditions, even if they did not reside there continuously. The fact
that these Kushite rulers took on Egyptian titles—Shabaka adopted the title
Neferkaure, Shabitku took the title Djedkaure, and Taharqa assumed the title
Nefertem-khu-Re—affords greater evidence yet of the strong embrace that
the Kushite rulers shared with Egyptian culture.17
In the first half of the seventh century B.C.E., the Assyrians wrestled
power away from the Kushites in Lower Egypt. Under the dominion of
probably only a small military contingency, the Assyrians allowed Neco I of
Sais, in the Delta, a certain autonomy of rule. In c. 661, the last Kushite ruler,
Tanutamun, sought unsuccessfully to regain control of Lower Egypt and the
Delta from the Assyrians. At his defeat, Psamtik I, the son of Neco, assumed
his father’s position as vassal in Sais to Assyrian ruler Ashurbanipal. Within
five years or so, Psamtik I had usurped the whole country, as Assyrian power
declined. Through marriage and peaceful agreement, Psamtik I consolidated
authority over Upper and Lower Egypt, and so began some 140 years of
recovery from foreign intervention—the Saite period (664–525 B.C.E.).18
The Saite period coincides with the flourishing of Greek colonization
and trade overseas. The employment of Greek mercenaries from Ionia, and
the founding of settlements in the Nile Delta, especially after 620, brought
Egypt into close contact with Greece.19 Consider the close proximity of
Naucratis to Sais, and the proliferation of monumental multicolumned temples
nearby in the Delta, represented in Figure 2.1.20
Generations of archaeologists and architectural and art historians working
in Greece, such as Tomlinson, Boardman, Braun, and Arnold, have expressed
the consensus of Egyptian influence on Greece. The contact with Egypt, an
ancient and advanced civilization, was of utmost importance for the Greeks,
and the consensus has been that Egypt contributed to the emergence and
development of monumental stone architecture in Greece. With the increase in
international trade, colonization, and mercenary assistance (especially from Ionia),
the thesis has been defended time and again that “a wave of Egyptian building
ideas, elements, and techniques infiltrated the Greek world during the twenty-
sixth dynasty.”21 Egyptian inspiration, then, motivated the enormous,
multicolumned stone temples of the Ionic and western Greek architecture of
the sixth century B.C.E. While the proportions of Doric colonnades correspond
to some degree to Egyptian examples, and the entasis of Doric columns has
strikingly close parallels in the swelling of Egyptian papyrus columns, the Ionic
proportions do not mirror so closely these Old Kingdom exemplars. However,
the enlargement of the central intercolumniation in Greek temples, Arnold
76 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
FIGURE 2.1 The Nile Delta in Egypt, Location of Temples, Saite Period
assigning, dividing, and certifying a standard (of quantity and quality). The
introduction of a new proportional rule for column height, 7:1, under the
Kushite rulers, suggests that a range of architectural innovations could be
accommodated; soon thereafter, the Ionian Greeks adopted a proportional
rule for column height, 9:1, or even 10:1, unlike the Doric 6:1 or 5:1. By the
sixth century B.C.E. in Greece, we have evidence for different interpretations
for the length of the cubit, and in the case of the tunnel of Eupalinos,
marking systems that indicate that, perhaps, Eupalinos introduced his own
tunnel measure.28 Thus in the course of the twenty-fifth and twenty-sixth
dynasties, the Greeks would have found, in the most ancient civilization of
which they knew, extraordinary and ongoing projects in monumental temple
architecture. They would have encountered a system that had sufficient
flexibility to permit innovations of all kinds while still embracing traditional
forms. And the Egyptians would have had, in turn, the kinds of cosmopolitan
input that might have increased further their own openness to apply innova-
tion to their traditions. It is from this architectural context that the innovative
dipteral Greek temples in Didyma, Ephesos, and Samos emerged. And it is
in this Egyptian and Greek context that Anaximander’s vision of cosmic
architecture attains its originality.
PROPORTIONALITY IN ANAXIMANDER’S
COSMIC ARCHITECTURE
In a recent book, Anaximander and the Architects,29 I argued that a significant
but unexplored source that fueled Anaximander’s cosmic imagination was the
archaic architects and their monumental temple projects. While acknowledg-
ing other important influences that likely affected Anaximander’s mentality,
Anaximander and the Architects sought to emphasize factors that had not been
discussed in the scholarly literature. As the architects worked in their own
backyard, Anaximander would have had the opportunity to observe their
methods and marvel at, alongside the awestruck members of his community,
their enormous achievements. He imported architectural techniques into his
cosmic thought because he imagined by analogy the architecture of the cos-
mos. This kind of reasoning by analogy played an important role in his
thought, and its importance has been discussed at length.30 To claim that
Anaximander had envisioned an analogy between the temple and the cosmos,
an analogy between the house of the cosmic powers and the house that is the
cosmos, however, is to acknowledge also the disanalogous elements in the
comparison. For, on the one hand, while the architects had to think carefully
about the circular column drums and the cylindrical forms the columns cre-
ated, archaic temple architecture is fundamentally right-angled. The rectan-
Robert Hahn 79
gular ground plan and the square plinths that supported the columns were
fundamental to the grid drawings by means of which the architects worked.
On the other hand, the cosmos that Anaximander imagined, and of which he
made drawings and/or model, is fundamentally circular and cylindrical. And
so, we must keep in mind as we proceed that while Anaximander imagined
the cosmos as cosmic architecture, he had to adapt and apply the architect’s
methods and technologies to a domain that was geometrically distinct.
Anaximander’s use of analogy is emblematic of his thought for it served
as both a mode of inference and a mode of discovery. He appealed to models
of human law (did¬nai gΩr a¶tΩ d√khn ka¥ t√sin ™ll–loiV), botanical
and embryological growth (g¬nimon, ™pokr√nesqai, ÷kkr√nesqai) and even
fish (Îcq§V) to provide the grounds for his inference and discovery of natural
law, organic growth, and human evolution. It is by appeal to certain kinds of
resemblance between things that Anaximander’s reasoning operates. And it
might be argued that analogy, for Anaximander, is a way to describe his type
of reasoning, reasoning from experience. While it seems clear that Anaximander
saw a resemblance between temple and cosmic architecture, we should keep
in mind that this resemblance does not preclude dissimilarity.
Anaximander had imagined the size and shape of the cosmos in the
absence of any adequate astronomical theory or instruments. He did so by
means of a variety of architectural techniques. Anaximander’s prose book, or
books, perhaps discussed many topics, but the discussion of the shape and
size of the cosmos deserves to be reviewed in light of the architectural
suggraj–.31 When the archaic architect and his patron agreed upon a temple
design, it seems likely that the details were written up. The “write-up” is
known by the Greek term suggraj–. The temple suggraj– was an ac-
count, in words and numbers, that detailed the architectural design, and it
served at least two purposes. First, it established a contract between patron
and architect, and second, it served as a reference for the building teams that
would undertake the various tasks of quarrying, transporting, installing, and
finishing the masonry. The section of Anaximander’s book that details the
structure of the cosmos, in words and numbers, mirroring the architect’s
s£ggraj–, is his cosmic suggraj–. He explained that the earth was of cy-
lindrical form, 3 1 in dimensions, just like a column drum; the heavenly
wheels of the stars, moon, and sun stand in increasing distance from the earth,
reckoned in earthly proportions—9/10, 18/19, 27/28. In each case, the first
number represents the distance to the inside of the wheel, the second number
to the outside of the wheel. Each wheel had a thickness of 1 earth-diameter.
Anaximander was not an architect of the cosmic house but rather a kind
of architectural historian of the cosmos. His cosmology offered a description
of the stages by which the cosmic architecture was created. While the archi-
tects designed and prescribed building techniques for cosmic and divine houses,
80 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
tral element in his cosmic architecture, and then, having adopted a modular
technique, he reckoned the size of the cosmos in proportions of it.
But how shall we explain Anaximander’s cosmic numbers? Lloyd recently
observed,39 rightly I think, that they were “symbolically appropriate,” but he
does not tell us how. The numbers were certainly not derived from astro-
nomical observations, nor were they derived by appeal to what we now regard
as astronomical theories. The usual recitation of those numbers has been 9,
18, 27, even though this series is just a conjecture.40 There certainly seems to
be something right about that conjecture, but to describe the series this way
tends to obfuscate rather than illuminate Anaximander’s method. The evi-
dence we do have, from tertiary sources, identifies the numbers 27 and 28 for
the sun wheel,41 and the number 19, not 18, for the moon wheel.42 No
numbers at all have been supplied to measure the star wheel, though we do
have testimony that the stars are closer to us than the moon or sun.43 When
the numbers 27 and 18 are assigned to the sun and moon, the star wheel is
identified with 9 to complete a mathematical series. The proposal of the
series 9, 18, and 27 arises from the conviction that Anaximander’s imagina-
tion is fundamentally geometrical.44 This view of Anaximander’s is, in part,
correct. But his geometrical mentality has been inadequately presented be-
cause we have not thought through the architectural analogy upon which he
was drawing, and in terms of which his audience could understand. When we
seek to grasp Anaximander’s geometrical proclivities, we must come to see
that his geometry is part of his method; that method attempts to illuminate
cosmic phenomena by appeal to proportions. Proportionality, like analogy, was
Anaximander’s method for making intelligible the astronomical phenomena
that exceeded his grasp.45
The numbers are commonly interpreted to give us either the size of the
wheels or the distance from earth to the wheels.46 I have already argued that
the numbers must be radii, not diameters.47 Anaximander is explaining the
distances of the stars, moon, and sun from us. He stands alone on the earliest
horizon of Greek thinkers who urged us to imagine depth in space, that the
moon lies immensely far beyond the most distant stars, and that the sun lies
immensely far beyond the moon. Unlike even his younger compatriot,
Anaximenes, who held that the stars, moon, and sun were like nails in the
crystalline sphere, and so all equidistant from us, Anaximander’s cosmic con-
ception highlights the varying distances to the stars, moon, and sun. The
uniqueness of his conception in the context of Greek meteorology is the
reason we should see that Anaximander was concerned principally with dis-
tances to, not the sizes of, the heavenly wheels.48
In defining distances, Anaximander also is following an architectural
technique. If we follow through with this analogy, we may reasonably conjec-
ture how he consciously worked. If this argument concerning Anaximander’s
82 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
marked out the distance from Europe to Asia to Libya on his terrestrial
map,53 and described the distance from each heavenly wheel to the other—
stars, moon, sun—on his cosmic map. Just as the architects made informal
drawings, with numbers and measures, and then turned to the stylobate and
made scratch marks to identify the exact locations for the installation of the
column bases and walls,54 so also Anaximander employed such techniques.
Where the architects and builders used a caliper, Anaximander likely used a
compass. Like the neo-Babylonian map, roughly contemporaneous with
Anaximander, the circular map might well have been made with a compass,
as has already been conjectured.55 When Herodotus ridiculed the early circu-
lar maps that he believed mistakenly identified the three regions of the earth,
as if they were all the same size, and as if they were made with a compass,
we have good reason to suppose that he is referring to Anaximander’s map,
at least by way of Hecateus and the Greek tradition, which Anaximander
both followed and subsequently inspired.56
Anaximander’s compass also replaces the architect’s caliper when making
his cosmic drawing. Anaximander imagined the cosmos in a horizontal cross-
section—that is, in a plan or an aerial view—as well as in orthographic views
other than plan, as did the architects when they imagined their houses of the
cosmic powers. When Anaximander made his cosmic plan—if he attempted a
drawing analogous to those he could have seen at the temple sites—he first
would have made a small circle to represent the earth and attached numbers to
identify its dimensions as being 3 times as broad as it was deep.57 Then, with
that compass, he would have counted out 9 earth diameters, just as the architect
would count out the intercolumniation in a basic unit of ancient measure, and
made a scratch mark. Next, Anaximander would have measured out one more
unit, the 10th earth diameter, since each wheel is 1 unit thick58—as defined by
the architectural module, the column-drum earth. With the compass, then, he
would have made two concentric circles, the first at a distance of 9 earth-
diameters, the second at 10, and he labeled his diagram with the numbers ‘9’
and ‘10,’ and thus he made the drawing of the first heavenly wheel. Anaximander,
we may suppose on this analogy, proceeded in a similar fashion for drawing the
wheels of the moon and sun. He would have repeated the same method for the
lunar wheel at 18 and 19 earth-diameters, and then finally for the solar wheel,
which stands the greatest distance from us, at 27 and 28 earth diameters respec-
tively. Consider Figure 2.2, in plan view. The claim that Anaximander followed
the architects by their design techniques helps us understand the process by
which his cosmic imagination was displayed. But will it explain the numbers
and proportions? As the reader can see, in this architectural light, the assess-
ment of Anaximander’s cosmic numbers as “9, 18, and 27” is woefully inad-
equate to grasp what he was doing and how he worked.
84 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
of conjuring immense distance, great amounts of time, and vast multitudes. Homer
calls upon the formula many times;60 so does Hesiod, who also invokes the
number 9 specifically in reckoning cosmic distances.61 In this light, Anaximander’s
series reads, on enormous scale, far, farther, and farthest.62 But we should keep in
mind that since “9” conjures a distance so great that it is rather inconceivably far,
the number expresses that immensity only symbolically. Thus while the stars, for
Anaximander, lie immensely far, the moon lies not merely farther but immensely
farther. So also, again, for the sun. The architecture that Anaximander’s numbers
summon us to imagine, therefore, is a staggering series of depths in space, each
immensely farther than an immensely far distance. We come to grasp its immen-
sity, if we do so at all, by its proportional relations.
Even in this context of archaic formulas, however, the description of
Anaximander’s series as 9, 18, and 27 is still inadequate, if not plainly mis-
leading. For Homer and Hesiod show us that the formula familiar to archaic
Greeks was not simply in terms of 9 but rather the combination of 9 and 10.
Homer makes Kalchas prophecy, in the Iliad, that the Trojan War will drag
on for 9 long years, but that the Achaeans will take the city in the 10th.63 In
the Odyssey, Homer makes Odysseus repeat that the sons of the Achaeans
fought for 9 years, and in the 10th year Troy was sacked.64 There are many
other examples of this formula in Homer.65 Moreover, the formula could be
doubled to show, proportionately, increasingly large epochs of human time, as
when Odysseus greets his aged dog, Argos, after “19” years (9 + 1 + 9), finally
coming home in the 20th (9 + 1 + 9 +1). The basic formula is still “9+1.”
The same formula is central to Hesiod. This is why earlier and important
studies of Anaximander by Gigon, for example, that place the emphasis on
the number “9” rather than on the combination of “9” and “10,” do not enable
us to grasp fully the archaic formula upon which Anaximander was drawing.66
The number “9” sometimes expresses a meaning when standing alone, as
when the poet refers to “9” swirling streams, that is, a great multitude of
them.67 But when Hesiod reckons cosmic distances, he tells us that the space
is so immense from the highest heaven to earth that the blacksmith’s weighty
anvil would fall ‘9’ days and nights and would reach the earth on the 10th;
so also for the distance from the gates of Hades to the lowest depths of
Tartaros.68 Hesiod’s reckoning of cosmic distances preserves cosmic symme-
try; the symmetry of Hesiod’s cosmos is illuminated through proportional-
ity.69 The formula is “9 + 1.” Thus, in Anaximander’s cosmic reckoning, the
distance to the stars is 9 + 1 earth diameters, not falling anvils; the distance
to the inside of the star wheel is 9 earth diameters and to the outside 10.
Anaximander’s moon is 9 + 9 + 1 farther from the earth than the stars, and
the sun is 9 + 9 + 9 + 1 earth diameters farther still.
There also seems to be operating in greater Ionia a canon of architectural
proportions, included among which is a rule for determining column height.
The architect-excavators, studying ancient temples, sought to determine
86 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
at the heart of his cosmic vision. In this case, archaic formulas that emphasize
“3” might well have underlined Anaximander’s inspiration to call upon it.
The seminal study on the ancient meaning of “3” was produced about a
century ago by Usener.77 In his essay, Usener assesses the importance of “3”
in Greek antiquity. In the first part of that essay, he notes the significance of
“3” and its continuations in various folk and religious traditions.78 He catego-
rizes various occurrences of “3” that may be regarded as most important,
focusing initially on the gods and expanding the parameters as he proceeds.
He counts fifteen different trinities of gods in Hesiod’s Theogony. He identifies
the repetitions of groupings that involve three gods or goddesses of equal
status. Next, he mentions a variety of triads that entail divinities of unequal
status. Usener then turns to consider patron deities, often in threes, the ap-
pellation of a trinity of gods in oaths, and the trinities of chthonic gods listed
along with places or temples, before exploring the proliferation and settings
of trinities among non-Greek peoples such as the Macedonians, Celts, Indi-
ans, and Egyptians, as well as the Christian usages of the triad. His surveys
attest to the fact that in archaic Greece, the symbolism of a triumvirate of
powers or principles was deeply embedded within the culture.
In the second part of his essay, Usener focuses on visual depictions of the
trinity.79 In visual forms, a physical “abbreviation” is common in which physi-
cal features are combined, as in gods who appear with three heads. The best-
known example from the Hellenic world is Hecate, who appears in numerous
works of ancient art, having a triune form—either having three bodies or
three faces. This also can be seen distinctly in the case of daemons, though
the depiction of three bodies with a single head was also common in Indian
and Greek traditions. Finally, Usener concludes this part of the investigation
by focusing upon cases where the gods appear either doubled or multiplied,
either as doubled with a third forming a trinity, or threefold accretions of
their own qualities. Thus not only conceptually but also visually, divine pow-
ers and perhaps the symbolic principles they embodied were familiarly rep-
resented in threes in Greek antiquity.
In the third and last part of his essay, Usener follows up by focusing on
the movement from “2” to “3.”80 God-pairs who become trinities are first
examined, then the division of the year (from two seasons to three), the
division of the heavens (from two winds to three), of paths (the cross in the
road becomes perceived as a three-pronged fork), and of the parts of the world.
In every case where it was physically possible, a progression was made from
pairings of two groups to three, so that even observations of nature were
subservient to this pattern; that is, regularities in nature were seen to conform,
or to be amenable to interpretation within, a threefold, rather than the more
familiar twofold, scheme. Usener lists, for example, linkages between duality
88 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
and double-sidedness, such as day and night, light and dark, up and down,
and right and left, and then observes that the use of the number “3” is not
apparently linked to any similar necessity. To describe adequately these tran-
sitions, he suggests that we use the phrase “mythological number system.”81
Usener characterizes further this number system as a “Pythagorean system” or
“arithmetic theology.” He argues that this later development misleads us,
however, into finding symbolism in numbers where they are what he calls
“screens” onto which analogies may be attached. Usener seems to be claiming
that, in the process of making experience meaningful, the ancient Greeks
organized and orchestrated their experiences in various typologies identified
by numbers. These numbers formed a system of background structures against
which, and on top of which, the details of our experience could be organized,
distinguished, related, and played out. Thus Usener urges us to see the for-
mative meaning that numbers had for ancient peoples, rather than the more
familiar view of grasping them—1, 2, 3—as merely establishing a sequence.
Usener then turned to single out numbers that attained a typological func-
tion. Large numbers, for example, were used for time concepts, starting with
the number ‘7’ and going up. In contrast, small numbers such as “2” were used
for expressions of quantity, while “3” was called upon to communicate complete-
ness. Perhaps within this conceptual frame, Anaximander’s tripartite earth should
be seen as just this breaking out of twofold pairings. On the one hand, the
accretion to a threefold, from a twofold structure is a breaking out from a
picture of the world embraced by traditional authority; on the other hand,
perhaps the threefold picturing is at once an expression of completeness, the
totality of the world, in a way that Anaximander and his contemporaries
found meaningful.
At the close of his study, Usener concludes that these usages of “2” and
“3” date originally from a period when no one among the Greeks could count
above three, and thus the concepts of perfection and fullness remained linked
to this upper limit.82 The idea that there was a time when the ancient Greeks
could not count above three may initially strain our credulity. Usener appealed
to the ethnographic research of Von den Steinen, whose work in central
Brazil at the close of last century offered a contemporary analogy.83 Accord-
ing to Von den Steinen, the members of the Bakairi tribe can only count to
“2,” and then continue “2 + 1,” and “2 + 2,” that is, they cannot count above
“3.” Usener concluded that there were still peoples in existence who had no
number concept higher than “2” and others for whom “3” was the highest
number and was equivalent to “a lot.” Locke, in the Essay,84 described the
arithmetical plight of American Indians in a similar vein; he insisted that they
could not count past twenty. It was not that they could not be instructed,
having reached a count of, let us say, twenty buffalo, but that the next in
Robert Hahn 89
the band around a door (= thyra) and has come to be called anathyrôsis by the
architectural historians.88 The rest of the drum face is countersunk, making
the inside concave, so that no part of the drum surface is raised above the
anathyrôsis band. This technique ensures the column’s stability; no part of the
drum face, except that band, will come into contact with the other drum. The
empolion is the other part of the solution for successful column construction.
It is a device employed to center the drum. After a hole was made in the exact
center of each drum, the empolion, a dowel usually made of wood, lined up
both drums, by joining both holes, as the new drum was lowered upon the
preceding drum or base. In earlier studies, the preparation of the drum face
was shown to bear a remarkable and suggestive likeness to a plan view recon-
struction of Anaximander’s cosmos.89 Aristotle had testified in the de Caelo
that Anaximander conceived the earth to be in the center of the cosmos
equidistant from all extremes [˛moi¬thta <t‹n g h̃n>.....pros–kei t¿ ÷p¥
toũ m°sou ˘drum°non ka¥ ˛mo√wV pr¿V tΩ ⁄scata ⁄con];90 the technique
of column-drum preparation required that the exact center of the drum face
be determined and that the extremes of the column drum, outlined by the
anathyrôsis band, like the “ocean” of Homer and Hesiod running round the
circumference, were also equidistant from the center.91 Consider the images
side by side in Figure 2.3. On the left is a rendition of Anaximander’s cosmos
(i.e., in plan or literally a horizontal cross-section through the plane of the
earth), and on the right is a reconstructed drum from the archaic Didymaion,
which just so happens to be roughly 3 1 in dimension and to have a round
empolion.
FIGURE 2.3 Anaximander’s Cosmos (Plan View), and Reconstruction of an Archaic Column
Drum. From the Temple of Apollo at Didyma, after Schneider
92 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
Now it might be objected that the drum face itself was an unlikely place
for Anaximander to have been cosmically and imaginatively inspired, because
the drum face would have become invisible upon installation. After the drum
had been made part of the column, no one, of course, would have seen the
drum face. And so an objection might be raised that if Anaximander imag-
ined the cosmos in terms of an architectural inspiration, why would he have
focused upon an element that no one could see once the column was erected?
The appropriate response might be that this architectural fact tends to sup-
port rather than to undermine the plausibility of the symbolic importance of
the drum face. Anaximander and the phusiologoi were trying to explain nature’s
hidden structure. They sought to reveal the underlying reality of things, which,
mere mortals though they were, they claimed the ability to penetrate. That
the drum face, whose appearance could have been known to anyone in his
Ionian community who visited the site during construction, became hidden
upon installation seems to support the arguments by analogy so pervasive in
early Greek thought.92 Anaximander’s archaic audience would have grasped
immediately the image of the drum face and recognized that, analogously, he
was claiming to truly reveal, or unconceal,93 an image of the cosmos. If, as it
has been argued in earlier studies, the column already had for his archaic
compatriots a cosmic and symbolic significance, the analogy would have been
transparent.94
There is, in addition, a new line of thought that further supports and
deepens our understanding of Anaximander’s terrestrial and celestial cartog-
raphy by architectural technology. This requires that we come to see the
architectural technique of trisecting, its application to the drum face, and its
indispensability for column construction. First let us consider why trisecting
was fundamental to archaic Greek architectural techniques. Among the basic
units of measurement, the ell (p–cuV) and foot (po§V) were central. Al-
though the exact lengths of the ell and foot have been the subject of great
debate among architect-excavators, and although the exact values of these
measures seem different at different locations and at different times,95 the
ratio of ell to foot is consistently 2:3. Dividing spaces into 3 equal parts, there-
fore, was fundamental to the systems of measurement in feet, and the ells constituted
of them. While 1 ell equals 1.5 feet, to avoid the fraction the relation could
be expressed simply in whole numbers as 3 feet = 2 ells. If we reflect on these
fundamental units of measure, we can begin to consider the geometrical
technique that must have been employed at the building sites. Figure 2.4
illustrates the common technique for trisecting: a line of any length can be
divided into three parts. The technique adumbrates theorem I.2 recorded by
Euclid at the opening of the Elements96—“To place at a given point a straight
line equal to a given straight line.” In this architectural technique, the geo-
metrical construction requires that a given straight line is divided in half, and
Robert Hahn 93
the length of the “half ” is transferred to the extremity of the given line. The
result is a line divided into 3 equal parts. Consider Figure 2.4.97
The trisecting technique was fundamental to archaic Greek architecture
because of the relation between ell and foot. But, the importance of trisecting,
FIGURE 2.5 Trisecting a Column Drum: Applied Geometrical Construction, after Hellner
96 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
3) USING A COMPASS,
THE DISTANCE IS SET AS
THE CIRCLE’S RADIUS, BC.
5) CONNECT CENTER OF
THE CIRCLE WITH POINTS
D AND E. THUS, THE
CIRCLE HAS BEEN
TRISECTED.
unmistakably that three marks were sufficient.99 The three-part marking system,
a consequence of trisecting the circular drum face, provided the technique by
which the architect could control the shaping and finishing of the whole
column. After the drums had been centered and placed one atop the other,
the stonemason would be able to finish the column, even when he no longer
98 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
had access to the measurements on the drum face, which of course had
become invisible upon installation. The continuation of the trisection to the
vertical sides of the drum, as the drawing with the workmen illustrates, made
this task possible.
The importance of three-part marking systems was fundamental to other
architectural projects in sixth-century Ionia. As Kienast showed in the recon-
struction of the tunnel of Eupalinos on Samos,100 contemporaneous with the
early stages of Dipteros II, a three-part marking system was necessary for the
architect to establish a straight line. One remarkable piece of evidence for this
appears in the south end of the Eupalinion. Only a short way into the south
end of the tunnel, one can look up and see a hole that goes to the top of the
hill. The purpose of it had challenged earlier speculations.101 Clearly, it was
too close to the south entrance to be a source of fresh air, and there were no
other such holes anywhere else in the tunnel, especially in the midsection.
Were the purpose of the hole for fresh air, surely the need would have been
greater toward the center of the tunnel. Kienast argued that, as the terrain
along the south end slopes downward, the architect could not get a siting of
three stakes that were needed to establish his line, since, as Kienast has
explained, Eupalinos originally had staked out the top of the hill to determine
exactly how long the tunnel was to be, by transferring this length along the
top of the hill into the tunnel. When the south end sloped downward and the
light from the entrance to the south tunnel was no longer visible, Eupalinos
actually brought the stake on top of the hill directly into the tunnel! It is for
this reason that the opening in the tunnel, upwards to the top of the hill,
appears shortly after entering the south end.
flourished. For, on the one hand, the Ionic architect’s conscious selections
echo patterns and aesthetic sensibilities that were meaningful to the Greeks
of the Archaic period and, on the other hand, these selections reified those
patterns by their monuments—they made “tradition”—and thereby estab-
lished a context that later Greeks would resonate, modify, and transform.
An exploration of architectural proportionality suggests that the temples,
for example, appear to have been conceived of as growing like a living
organism.102 The creation of the buildings seems to have been consciously
planned in such a way that we can say that the architects allowed the
building to grow. Our explanations might begin, as did Schaber’s, with wall
width.103 Then, the axis width is determined in terms of wall width. From
the axis width, the temple often “grows” in terms of that basic unit to
determine the length of the cella. Then again, by the same increment or a
multiple of it, the temple grows to create the sekos or inner built structure
that is sometimes referred to as the “cella building” or “Kernbau.” Then, by
addition of the incremental unit, or in terms of it, the temple grows to
include the peristasis. Sometimes it seems that the architects allowed the
sacred temple to grow in such a way that the distance to the altar, the focus
of the sacrifice so central to their religious rituals, also was calculated in
terms of the same basic increment.
In this “organic” analogy, the temple, like the earth and the cosmos, is
conceived of as a living and growing being. The whole sanctuary is alive, in
both its natural and created environment. In imitation and reflection of this
reality, the temple design expresses this organic character. In Samos and
Ephesos, the great archaic temples were built in marshes, alive with all man-
ner of living creatures—frogs and birds, insects and vegetation, animals of
great variety—sacred to the fertility goddesses Hera and Artemis. In Didyma,
the huge hypaethral temple contained a spring and a grove of laurel trees
sacred to Apollo. In these sites, selected especially because of their sacred
celebrations of nature, it should come as no surprise that the temple designs
were conceived of organically. Indeed, the temples celebrated the divine pow-
ers, the living and fecund powers of nature.
This organic vision of nature is familiar in archaic Greece. We have
evidence, for example, that Anaximander imagined the cosmos as an evolving
thing; this is fundamental in understanding his cosmic imagery. Sometimes
couched in embryological terms, and at other times in botanical terms,
Anaximander’s cosmology invites us to imagine the universe as a living or-
ganism. On the one hand, we have reports of Anaximander’s cosmology
couched in terms such as g¬nimoV, ™pokr√nesqai, and ™porr–gnusqai
terms that have technical senses in embryology.104 On the other hand, we
have evidence for terms with a peculiar botanical sense such as d°ndron and
100 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
jloi¬V. The imagery of the tree and bark points us in several directions: it
invites us to consider the role of sacred trees in cosmological discussions,105
the centrality of a cylindrical rather than a spherical model,106 and it brings
us back directly to the temple and early sacred architecture in the construc-
tion of which trees played an essential role.107 The idea of the cosmos as a
living organism, however, is a distinctive view of Anaximander and the early
philosophers. Indeed, it is emblematic of the origins of philosophy, that all of
nature is perceived as being alive—and in just this sense is a cosmos.108 But this
hylozoism seems to arise as an extension of the belief that individual phe-
nomena are alive. The archaic background is suffused with beliefs that natural
phenomena, on the one hand, are individually alive, and on the other hand,
are the result of biological generation.
Homer and Hesiod offer abundant evidence to suggest that natural
phenomena were conceived as living beings or generated beings. While we
have no positive evidence in the surviving writings attributed to them that the
cosmos itself was perceived as a living creature,109 various natural phenomena
in it certainly were treated as if they were alive. In Homer, Heaven and Earth,
the rivers, winds, and sun are invoked to witness oaths or are addressed in
prayers, as if they had the power to influence events.110 In Hesiod, we learn
how the gods, and the earth, rivers, the immense sea, shining stars, and
Heaven, all came into being.111 In Hesiod, while some phenomena are ac-
counted for by parthenogenic reproduction, as when Earth brings about Heaven
without delightful love,112 the admixture of Eros accounts for most genera-
tion. Thus the generation of nature is broadly conceived of in biological
terms, whether sexual or not. Because they are generated, the parts of nature
are alive. Anaximander appears to have extended this reasoning; since all of
nature is generated, it too must be alive. Consequently, in the context of
archaic culture, it would not be surprising to discover that the temples were
conceived of organically. The temple imitated and reflected the life-form; the
temple was the sacred dwelling of the god, and its design was appropriate to
the living beings who would come to dwell therein.
excavator has been to determine the ancient units of measure that were
employed in the design and construction of buildings. How does one work
backwards, as it were, to figure out the basic increments in terms of which the
building was erected? Hultsch set out the problem by casting it in the shadow
of Protagoras’ dictum, that “Man is the measure of all things.” In this asser-
tion, Hultsch regarded the foundation of all systems of measurements.114 All
measurement is a comparison, and a certain dimension is made to serve as a
basis for a system and is used as a measure for similar dimensions. At first,
the human body seems to have served as the object of comparison for spatial
dimensions. It was only a small step from the use of natural measurements
to the introduction of artificial measurements, created according to an agreed-
upon norm. Without such a norm, architecture would be unimaginable. For
this reason, Hultsch pointed out that the oldest masters of building in the
world, the Egyptians, also have the oldest standard of measurement, and
these units of measure initially were consequences of the need to measure the
land following the annual inundation of the Nile—the natural origins of
geometry.115 The direct sources for the metrology of ancient peoples are the
measuring sticks, containers, weights, and coins that are still in existence, but
with the exception of coins, for which we have numerous examples, few of
these other standards survive. The few foot-measuring sticks or carvings in
stone that do survive do not give us a reliable value for the foot that they
depict and, in addition, it is for the Roman foot, not the Greek foot, that we
have this evidence.116 Moreover, in order to determine exactly the ancient
units of measure—which certainly seem to have varied from city to city, and
from time to time within each city—we must keep in mind certain complexi-
ties of the problem. The ancient weights and measures that did survive were
not standardized according to mathematical precision, but, since they were
intended for practical use, they provide only an approximate picture of the
normal measurement. The problem can be grasped in a similar fashion in our
contemporary world; even though we are much more exact in such matters,
it would be impossible to derive the normative system of measurement from
measuring devices in everyday use. For example, from the measuring cups in
our modern kitchens, or the rulers in our desk drawers, it would be enor-
mously difficult to derive the rule for constructing these units of measure.
This situation is to be expected, and even more so, in ancient measurements
where the conditions were even less favorable. So how is it that we can
recapture the ancient units of measurement with which the architects worked?
The Greeks did not develop their measurements independently but rather
adapted and refined the systems of measuring lengths and surfaces that they
inherited from Egypt and the Orient. The basic units of measure—ell (p–-
cuV), foot (po§V), dactyl (dºktuloV)—in Ionia seem to have been adapted
from these sources.117 Homer knows of the po§V,118 p–cuV,119 and o’¢rguia120
102 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
Pyramid of Cheops, for example, the dimensions of the building were mea-
sured and then divided by the approximate value of the King’s ell. The result,
by simple arithmetic, was that each side was exactly 440 ells, on the condition
that the value of the Royal or King’s ell was 0.523 m. In the seventh century
temples to Hera in Samos, where an early one was replaced by a later one of
the same length (though the second was of greater width), the excavators
concluded that the Heraion was exactly 100 Samian ells in length and, con-
sequently, referred to as a Hekatompedon or “hundred-footer.” When the de-
cision was made, sometime prior to 575 B . C . E ., 141 to build the first
double-pteron Heraion completely out of stone and on an immense scale, the
so-called Dipteros I or Theodorus temple, the platform and the stylobate, of
course, were enlarged greatly. According to the reconstruction of Dipteros I
by Schleif, the dimensions were 200 Samian ells long by 100 Samian ells
wide.142 The basic proportion was 2:1: the temple was twice as long as it was
wide. Since the front elevation is conjectured to be roughly 50 ells in height,
the basic proportions were achieved by a principle of simple doubling. Dipteros
I was twice as wide as it was high, and twice as long as it was wide.
When we think about the metrology of ancient buildings, however, we
must distinguish between the plan and the actual building. In the Samian
Dipteros II, the so-called Rhoikan or Polycrates temple, the one that replaced
the Theodorus temple (= Dipteros I) that was destroyed sometime after the
middle of the sixth century, the excavators claimed that the temple plan was
different from the actual building that resulted. How are we to understand
such a claim? Reuther argued that the plan of the sekos, that is the cella with
the pronaos, and which we shall call the Kernbau, is about 1:3, which is 52
156 ells, but the actual length was shorter by 1 ell.143 The total length of
Dipteros II standing in relation to the total width, like Dipteros I, could be
seen in the plan in 2:1 proportions, but the actual measurements in ells are
207144 100. A similar, related problem is evidenced in the pronaos where,
according to Reuther, the plan was shortened by 1 ell in the actual building.
Reuther had no answer for this discrepancy, nor, apparently, do the excavators
who followed.145 In any case, we can make sense of the conceptual strategy.
The excavators first grasped the general proportions and then, reflecting upon
their measurements in terms of the approximate sizes of ancient ells and feet,
they reached the conclusion that the architects had consciously set out the
sixth-century temples to Hera in Samos on a plan of 200 100 ells.146
Having determined the ground plan, the actual building measurements varied
slightly from the plan. The proportions were judged unmistakably to be 2:1,
and that pattern came to be regarded as the plan. And since the particular
measurements were expressed in numbers so close to 200 100 ells, the
excavators were compelled to conclude that its plan was 200 100. Thus, by
104 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
this distinction between plan and actual building, Reuther was able to con-
clude that the ell is 0.524 m because the eastern temple front is 52.45 m and
there was assumed to be a plan of 100 ells in temple width.
From all of this, we might describe the architect-excavator’s sleuthlike
technique in terms of a hermeneutic circle. In the hermeneutic circle, we at-
tempt to interpret a word, whose meaning we do not know, by appeal to the
context in which the word appears. But of course we cannot be sure of the
context without knowing the meaning of the crucial word in question. So,
each time we go around the circle—from context to word, and around again—
we reflect on the word as we find it, and we discover a little more to illumi-
nate the context in which the word occurs. As the context is illuminated
increasingly, the meaning of the unknown word becomes clearer, little by
little, thus further illuminating the context, and so on. Analogously, the ex-
cavators determine ancient units of measure: they examine and reconstruct
the surviving dimensions of ancient buildings and then try to abstract the
proportional design expressed; then, on the presumption that the Greeks
reckoned their constructions in “round numbers”—using measured cords and
ropes—the excavators divide the lengths by various suggested ancient assign-
ments until they come to see both the plan and the actual building display
a basic unit. This metrological technique operates by a form of coherence
argument.
These “round numbers” lead us to another important point in understand-
ing ancient metrology and the architectural design techniques that are presup-
posed. The idea that the ancient architects designed their monumental buildings
by appeal to complex mathematical formulas has little support. This does not
mean that the selection of numbers and proportions expressed in terms of them
was arbitrary. Instead, it seems that specific numerical assignments were ini-
tially chosen, perhaps for reasons that were symbolically or formatively appro-
priate, although within a range of supposed structural tolerances.147 Then, by
these measured cords, the dimensions of the building were marked out in whole
or round numbers.148 The main point is that the archaic Greek architects adopted
whole-number intervals, and simple proportions, whenever possible. Stated
another way, the archaic architects did whatever they could to avoid complex
fractions. Among historians of mathematics, the familiar view is that Greeks
followed the ancient routine of avoiding, by the use of submultiples, the difficulty
of computing with fractions.149 This seems to have been especially true in
archaic architecture, when monumental stone building in Greece was in its
early stages, though the evidence that we have from the fifth and fourth cen-
turies is lacking in the sixth century. From the evidence that we do have, in
general there seems to be no numeric notation for fractions applicable to archi-
tecture, and the architects seem to have favored submultiples, with a preference
Robert Hahn 105
for small numbers as denominators.150 Coulton observed that the general ten-
dency of Greek architects was to work with sizes that could be expressed in
round numbers of feet or simple fractions of feet, that this tendency was part
of their design technique, and that rounding out of dimensions would clearly
simplify the architect’s task, although it complicates the work of modern inves-
tigators of Greek design.151 The complication arises, of course, when we try to
infer the design, when the actual measurements, due to rounding up, tend to
occlude rather than illuminate the metrological plan. This is just the compli-
cation that challenges the excavator when the metrology of the actual building
varies from the supposed plan. Coulton recommended that the study of archi-
tectural proportions should, theoretically at least, precede the study of units of
measurement,152 and as we discussed in the case of Reuther in Samos, the
architect-excavators working in Ionia seem to follow just this procedure.
Wesenberg’s Case for the Canon of Ionic Proportions: 1:9 not 1:10
UD
ud Apophyge
Plättchen
FIGURE 2.9 The Debate over where on the Column Drum “Lower Diameter” is to be
Measured (after Wesenberg)
Then, having assigned “ud” the role of the module, equal to 1 unit, the
other proportional relations are calculated in terms of it. With this definition
of the module, Krischen began an examination of the exemplary masterpieces
of the late-classical Ionian architecture, and working backwards, as it were,
sought to infer what had been the rules of proportion for the Ionic Order.
These rules have come to be known as the Ionic Canon. Krischen concluded
that when the entablature has no frieze, the lower diameter of a column
equals 1/10 of the column height; the height of the entablature is 1/6 of the
column height; and the architrave is in turn 1/2 the height of the total
entablature.160
A cornerstone of Krischen’s thesis is that the fundamental canon of pro-
portions is 1:10; the module is identified with “ud,” and is equal to 1 unit, and
the column height is calculated, in the absence of surviving examples in situ,
to be 10 times this unit diameter. When this fundamental canon is accepted,
Krischen elaborates on the canon in terms of the proportional relations among
key architectural elements, namely, the “lower column diameter” (ud), the
“column height (SH),”161 the entablature height (GH),”162 and the “architrave
height (AH).”163 The various heights of these key architectural elements stand
in simple proportions, amenable to practical measurements. By Krischen’s
reckoning, the Ionic Canon can be expressed by the following interrelations
in table 2.1:
108 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
TABLE 2.1
The Ionic Theory of Proportions, after Krischen
the defense of this new formulation rests on the presumptions that: (1) Vitruvius’
work suggests that “UD,” and not “ud,” is the module of the measurements; (2)
1:9 has meaning beyond individual cases; and (3) the new formula leads to a
practical source for the derivation of all other sizes, unlike Krischen’s formulation.
TABLE 2.2
The Ionic Theory of Proportions, modified by Wesenberg
Wesenberg, however, is not dogmatic about the case that he has fash-
ioned. He acknowledges that either view—1:9 or 1:10—could be defended,
and although he has proposed reasons why 1:9 seems to offer more practical
numbers to more Ionic buildings, he acknowledges that the case cannot be
made with certainty. So what can we conclude about the context in which
Anaximander made his assessment of the dimensions of cosmic architecture?
The column was the defining feature of the Greek temple. In the sixth-
century B.C.E. double-pteron Samian Heraion, Ephesian Artemision, and
Didymaion, the column extends more than 50% the total height of the eleva-
tion. In this further sense, it was the defining feature of the visual experience.
Wesenberg’s case study confirms again that a proportional rule was operative
in which column height is “9” or “10” times the lower column diameter. And,
in agreement with Vitruvius, the column diameter was the architect’s module
for determining column height. The column symbolically separated Heaven
and Earth, and in Ionia, “9” or “10” times the diameter of a (lower) column
drum was thought to express symbolically or formatively the appropriate
distance. Anaximander’s conscious selection of 9/10 earth diameters to locate
the beginning of Heaven shares the same architectural rule of proportion.
us unravel the simple ancient proportions with which the ancient architects
worked and help us grasp both the general patterns and the metrological
units of the archaic Greeks. The case that ancient architects envisioned their
temples as growing organisms, allowed them to grow according to certain
formulas, and rooted their constructions in a conscious selection of numbers
can be made by appealing to a metrological study of ancient buildings. To
explore this thesis, we begin with Schaber’s provocative study of the seventh-
and sixth-century B.C.E. temples of Artemis in Ephesos—Artemision “C”
and Artemision “D”—and then we compare and contrast the metrologies of
the contemporaneous temples in Samos and Didyma. When we have finished
a cursory metrological inspection of these great monumental stone temples,
new to Ionia at just the time when Anaximander flourished and “published,”
we will then return to Anaximander’s proportions and numbers, for a second
look, in light of what we have discerned from the metrological studies of
Wesenberg and Schaber, and the additional architectural formulas and met-
rological analyses of Schneider’s study of archaic roofs in Didyma.
In ancient Ephesos there was a monumental stone temple dating to the
late seventh century, but apparently not dipteral. This temple, generally re-
ferred to as Artemision “C,” predates immediately one of the most famous
temples from Hellenic antiquity, Artemision “D.” Artemision “C” is known
also as the “Pythagoras temple”; it dates to the late seventh century, and it is
identified with the time of the tyrant Pythagoras of Ephesos (not the sixth-
century philosopher from Samos). Artemision “D” is known also as the “Kroisos
temple,” and dates to the mid-sixth century, perhaps as early as 560 B.C.E.,
soon after Kroisos, a major patron, ascended to power in nearby Lydia.168
Known as one of the seven wonders of the ancient world, the Kroisos temple
was made entirely of marble. The Pythagoras temple and the Kroisos temple
both seem to have the same “number” structure, but the Kroisos temple was
significantly larger, since its proportions were reckoned in ells, not feet. While
the ground plans follow the same initial numerical pattern—axis width, cella,
sekos—44, 88, 132, that is, an increase of 44 in the basic units of growth, the
earlier temple, Artemision “C” is comprised of feet—44, 88, 132, while
Artemision “D” is measured in ells—44, 88, 132. And since an ell is 1.5 feet,
the temple ground plan is significantly larger and is made to grow, as it were,
in a geometrical, and not an arithmetical, series. Consider Figures 2.10 and
2.11 of Artemision “C” and Artemision “D,” after Schaber,169 and Table 2.3
that follows, showing how Artemision “D,” the Kroisos temple, was much
greater in size. Thus the ground plan measurements for the seventh-century
Artemision “C” display a progression of an arithmetical series with an in-
crease in each case of 44: 44, 88, 132, 176, and 220. Schaber describes this
strategy as a “doubled doubling” (doppelte Verdoppelung). He explains the roots
of the series by reducing the components, twice by halving, to: 22, 44, 66, 88,
and 110; and then 11, 22, 33, 44, and 55; and finally to 1, 2, 3, 4, and 5, to
Robert Hahn 111
FIGURE 2.10 Temple C [Late Seventh Century B.C.E.] of Artemis at Ephesos and Temple D
[Mid-Sixth Century B.C.E.] of Artemis at Ephesos (after Schaber)
reach the lowest common terms of the series, where the “Pythagorean” (=
Pythagoras of samos) number series of the sacred tetractys, 1 + 2 + 3 + 4 =
10, is being emphasized: 10 + 1 = 11, and so on. Schaber proceeds this way
to argue that: (1) the basic numbers that were selected for the building elements
had a formative power; (2) the buildings were made “to grow” in terms of these
formative numbers; and (3) the selection of these basic numbers reflects the
importance of the use of the abacus for the architect’s imagination.
The dimensions that constitute the structure arise from a numerically
strict arithmetical series of a doubled doubling. Schaber invites us to think
about the kind of notation in which such a series would likely have been
represented, and thus how the archaic Greeks in Ionia would have thought
about such numbers. In the so-called early Herodian number script, the
112 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
BREAKDOWN
CELLA
& AXIS WIDTH
44'
44' 88'
132'
DOUBLE
PERISTASIS
176'
COMPOSITE
FIGURE 2.11 Temple D of Artemis at Ephesos—the Proportions and Numbers of the Temple
Parts (after Schaber)
Robert Hahn 113
TABLE 2.3
Metrology for the Ground Plan of Artemision “C” and “D”
number sequence 11, 22, 33, 44, and 55 would appear in the mid-seventh
century in the simple images: DI DDII DDDIII DDDDIIII DDDDDIIIII. In terms
of these images, Schaber points out that it would be easy to imagine a y›joV,
a doubling of the initial series 1, 2, 3, 4, and 5 as it could be displayed on an
abacus, in the mensa Pythagorica. This imaging he regards as being determi-
native of their selection.
• • 11
• • 22
• •
• • 33
• •
• •
• • 44
• •
• •
• •
• • 55
• •
• •
• •
• •
What these images suggest, then, is that if an archaic Greek wished to ex-
press the simplest starting point for the series 11, 22, 33 . . . , the series could
be broken down into “10 + 1,” “10 + 10 + 1 + 1,” “10 + 10 + 10 + 1 + 1 +
1,” and so on. This is another way of expressing that “10” occupied a special
position at the end of the number series;172 with 11, the series starts again,
114 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
so to speak. The number “10” also enjoys a privileged place among the
Pythagoreans. The sacred tetractys, considered by the Pythagoreans as a source
and root of all j§siV, identifies 10 as the sum of the series 1 + 2 + 3 + 4. As
Schaber put it, “the productive power of ten is the number four.”173
The proportions of the ground plan of the Kroisos temple, Artemision
“D,” are identical with those of the Pythagoras temple, Artemision “C”—the
basic changes, metrologically speaking, result from the reckoning in ells, not
feet. While the geometrical series 88, 132, 198, and 297 does not continue
the incremental additions in Artemision “C,” the pattern instead refers back
to the previous dimension, being multiplied by 1.5. Schaber then expresses
this relation in such a way as to replace “times 1.5,” in the conversion of feet
to ells, with the expression “plus a half.” The “plus a half ” corresponds to the
ratio 3:2, the mathematical midpoint between 2 and 1. The principle of
doubling, and halving, affirms a female principle of generation, identified
with Artemis/Kybele; this reflects an “organism,” or organic growth.174 We
have already considered in the discussion of metrology that the fundamental
relation of “foot” to “ell” is 3:2. The relations of metrological units, such as
they are, lent themselves to these patterns of expression.
Schaber’s theorizing about the organic growth of the temple focuses more
on the ground plan than on the elevations, since the evidence for the ground
plan is more certain.175 However, the senses of proportionality—the basic pat-
terns and rhythms of the buildings—are communicated to the archaic commu-
nity most directly through the elevations. How did the buildings “feel” to the
local inhabitants as they came into view of the great temples? What basic
patterns predominated? What was the relation between the height of the build-
ing and its width? What was the relation between the height and its length?
What was the relation between the height of the column and the height of the
entablature that stood on top of the capital? Schaber also addresses these prob-
lems in his reconstruction.176 His thesis is that in Artemision “D,” the Kroisos
temple, the total elevation of the building is exactly equal to the axis width, that
is the width of the cella, 44 feet. This conscious selection of numbers harkens
back to the abacus where “44” was expressed as:
Tens Ones
• •
• • 44
• •
• •
According to Schaber’s reading, the selection of “44” was a conscious expres-
sion of the productive force of nature, based on “10” the sacred tetractys, and
“4” its constituent parts.
Schaber’s reconstruction of the elevation begins with the plinth that stands
directly upon the stylobate. The elevated courses of construction then include
Robert Hahn 115
the torus, column, capital, architrave, kyma, dentils, geison, sima. Consider
Figure 2.12, after Schaber.177
FIGURE 2.12 Artemision D: The Kroisos Temple Numbers and Proportions of the Elevation
(after Schaber)
116 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
TABLE 2.4
The Metrology of the Elevation of the Archaic Temple of Artemis in Ephesos
cal, and does not remain simply arithmetical, like the Pythagoras temple, as
the series continues with cella, sekos (Kernbau), peristasis, temple + altar of
88, 132, 198, and 297. Each subsequent size may be reckoned, Schaber points
out, by multiplying the previous number by 1.5. This is the same as calculat-
ing each subsequent dimension by “plus a half ” and equals the proportion of
3:2, an expression of the basic relation of foot to ell. Thus Schaber concludes
that the number of feet in the cella length, the number of ells in the sekos
(Kernbau) length, the sekos (Kernbau) measurement in feet, the peristasis in
ells, the peristasis in feet, and the length of the temple and altar in ells, were
all produced by the relationship of F:E = 3:2.
88 + 44 = 132
132 + 66 = 198
198 + 99 = 297
concludes that the series for both temples reinforces the idea that principles
of doubling and halving are the most important productive principles of
number, and ultimately the origin of all things made of number.184
In the assessment of cosmic distances, Anaximander reckons that the
nearest point to the stars is 9 earth diameters from us. The nearest distance
to the moon is exactly double that distance. Following an arithmetical pro-
gression, beginning with the distance of 9, the sun stands another increment
of 9 further than the moon. Anaximander’s cosmos, too, certainly can be
described as growing in basic increments. In terms that Schaber supplies to
describe the “growth” of Artemision “C,” datable to the late seventh century
or early sixth century B.C.E., the architectural development of Anaximander’s
sacred and cosmic house, the one in which we all dwell, follows an arithmeti-
cal progression. However, if we approach these measures, as Anaximander can
be argued to have done, beginning with his module—the number “3”—then
a new proportional pattern of cosmic distances appears: 3 3 1, 3 3
2, 3 3 3. This is not the whole story, but it advances our search for
his formula. To see the formula that Anaximander seems to have adopted,
comparable to an architectural design formula in use in sixth-century Miletus/
Didyma, we must wait until we have reviewed Schneider’s metrological study
of a cult building there. For now, we turn to compare the metrologies of the
great archaic temples of Hera in Samos and Apollo in Didyma, to the temples
of Artemis in Ephesos, discussed by Schaber. When we make that compari-
son we shall discover that the Kernbau is usually, but not always, in 3:1
proportion. Anaximander’s module, his column-drum earth, is identified by
its 3:1 proportion. Furthermore, if we identify the distance from the earth to
the nearest heaven of the star wheel, and call that “1” unit, the moon and the
sun stand more distant from us by “2” and “3” units. Did Anaximander
envisage the cosmos itself on some sort of analogy with the temple’s sacred
Kernbau, its inner built structure? While the source and origin of all things
for Anaximander is the Òpeiron, the unbounded or unlimited, the cosmic
structure from farthest sun wheel to the earth presents us with a limited and
bounded, inner built structure. Had Anaximander imagined the cosmos as
cosmic architecture, could temple design have encouraged him to choose, by
analogy, the proportions of 3:1 as did the architects for the temple’s inner
built structure?
In Schaber’s study, the dimensions of the ground plan and the elevation are
connected. A central assumption of the work of architect-excavators, of course,
Robert Hahn 119
is that the whole building is thought through prior to the construction, and
the dimensions of the parts stand in relation to one another according to
canonical, proportional rules. This means that the numbers and proportions
of the ground plan are inextricably connected with the numbers and propor-
tions of the elevations. While the ground plan usually is much better pre-
served than the elevated parts of the building, which now lie in ruins, the
better-preserved ground plan is regarded to hold the key to the reconstruction
of the elevated dimensions. By the more reliable evidence for the ground
plan, Schaber reconstructs the three-dimensional views of the temple. Of
course, the visceral and optical impact of the building is delivered in the
three-dimensional experience of it, that is, the experience that a person has
upon approaching the building and walking its perimeter, and only indirectly
by the pattern established in the ground plan. In Schaber’s reconstruction, the
height of the elevation of the Artemision from the plinth to the roof is equal
to the length of the inner built structure, or Kernbau.185 And both have a
measure of symbolic importance, namely 44 feet. Thus the inner built struc-
ture holds a special place in revealing the fundamental design patterns of the
temple.
When the metrologies of the Dipteros I and Dipteros II of Hera at
Samos, both datable to the sixth century, are placed alongside those of the
archaic temple of Apollo in Didyma, certain fundamental patterns can be
seen; these patterns are illuminated further, when comparing these temples
with the archaic Artemision. A comparison of the metrologies of axis width
and Kernbau shows that all the buildings, with the one exception of the
Didymaion, exhibit the general pattern of incremental growth—1:2:3. The
Heraion datable to 575 B.C.E. grows in the series of 50, ?, and 150—but the
peristasis length of 192 makes likely an assignment of approximately 100 to
the cella length about which the evidence is less certain; the slightly later re-
built Heraion grows in the same pattern 49, 103, and 155 (and then peristasis
207). The Didymaion seems to be the exception to the general rule but the
pattern is similar. Gruben suggested the pattern of growth to be 50, 144, and
198, that is 1:3:4 and not the usual 1:2:3.186 More recently Fehr has conjec-
tured, from the uncertain evidence, that the series might instead be 48, 96,
and 192, that is, 1:2:4.187 In the cases of both the Heraion and Didymaion,
the buildings can be described as growing in terms of the axis width. But
both display the same proportional patterns.
Let us consider the metrology for the Samian Heraion in table 2.5 and
by comparison the archaic Didymaion in table 2.6. From these metrologies,
we can see that the monumental temples in Samos shared the same overall
architectural proportions as the temples in Ephesos that Schaber detailed.
One way to describe the measurements is to say that the temples were made
to grow in terms of a basic unit, namely, axis width. The axis width, Schaber
120 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
TABLE 2.5
The Metrology of the Archaic Temples to Hera in Samos
TABLE 2.6
The Metrology of the Archaic Temple of Apollo in Didyma
Everyone who reads the orally composed epic poetry of ancient Greece is
struck by repetitions. These repetitions appear as words, phrases, portions of
verses, or even groups of verses. The reader now, as presumably the listener
then, is struck by the reappearance, again and again, of familiar refrains. No
doubt the use of these formulas, these packaged expressions, would be a great
aid to the poet as he sang his lengthy songs. But the appeal to practical
advantages alone cannot explain the existence of formulas. Just as the archi-
tect could modify and refine the details of temple design, so also could the
epic poet; but, just as the architect did not wish to wander too far from
conventional patterns as he reaffirmed the permanent and eternal form of his
vision, so also the poet. The originality of the poet, and the architect, then,
did not require new wrappings for the time-tested patterns; rather, both
resorted to them, and by means of them, as they attempted to present a
content ideally suited to its form.
The epic poet gave expression to his vision in verse. Since style and
content were inextricably bound up with the structure of the verse, intellec-
tual analysis of the technical details seems to fall short when we try to attain
122 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
crackling moment, the fire is changed. In the universal process, he says, there
is a constant exchange of fire for all things, and all things for fire, like goods
for gold and gold for goods. In the Greek expression crusoũ cr–mata ka¥
crhmºtwn crus¬V, perhaps the ancient audience heard the crackling of
fire.192 Or in Heraclitus’ expression potamoĩV toĩV a¶toĩV ÷mba√nom÷n te
ka¥ o¶k ÷mba√nomen, we can hear the river’s water splashing and swirling as
we step and do not step into it.193 Poetic meaning might certainly be con-
veyed in a variety of ways and on a variety of levels, but it is useful to invoke
a distinction between literal meanings in terms of which a story unfolds, and
the syllabification that, while meaningful, is perhaps not the specific conveyor
of this kind of literal meaning. Thus in epic verse the content is delivered in
packages of long and short syllables that, in themselves, are perhaps not, as
whole words and phrases would be, the literal carriers of meaning. Seen in
this light, the problem for the ancient poet was to find a compromise between
two extremes. Either the poet could try to deliver a meaning, as in ordinary
speech, by emphasizing words or phrases, and thus run the risk of losing the
rhythm, or the poet could continue the rhythmic patterns and forego the
effective delivery of the literal meaning.
The poet’s compromise, then, might best be seen to be realized not on
the level of the syllable but rather on the level of the verse. Hexameter verse
runs rhythmically in six short elements of two or three syllables each. Indeed,
the hexameter is a succession of six measures or feet. Each foot measure is
either a dactyl, that is, a “finger”—long, short, short, like the articulation of
fingers, or a spondee, long-long. The last foot in the line consists of two
syllables: long-short, or long-long, and the fifth is almost always a dactyl.
While the rhythm is delivered in elements of three or two syllables each, the
meaning is delivered in verses, fifteen to seventeen syllables each, divided here
and there but not rhythmically determined, except for the last two feet. The
poet’s success in hexameter verse, then, is achieved on the level of verse, not
syllable; by dividing each verse into four segments or cola, the poet’s meaning
is transmitted while allowing the rhythmic patterns to continue the flow. The
first colon is placed at the beginning of the verse, and the three succeeding
cola commence with a perceptible caesura or interruption. The poet can in-
troduce the first caesura at four different points, while the second and third
interruptions can occur at two different moments within the verse. The result
is that the poet can avoid the monotony that would naturally result if the
verses always followed identical patterns, and yet the varieties of verse pat-
terns remains simple and uniform. The reader, or listener, begins to sense the
rhythm of the verse, and the rhythm of the meaning, soon after a few verses
have been delivered. In all of this, intellectual analysis was neither useful nor
necessary for the appreciation of hexameter verse. No faulty verse could be
124 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
recited in the traditional manner without being noticed, and the system of
rules was discovered implicitly through the process of hearing, and perhaps
reciting, the hexameter.
The situation is analogous in temple architecture. The size of the blocks,
like the poet’s syllables, was not the primary communicator of the meaning
of the temple, though both were reckoned in feet and dactyls. While the size
of the blocks, like the syllables, usually was not a carrier of meaning, together
they formed conglomerates that were. Consider an architectural example.
When temple building became monumental and relied upon stone, not wood,
for supporting the heavy entablature and roof, monolithic columns, for ex-
ample, were replaced by stone drums to solve the difficulties of delivery and
installation of these weighty components. The size of the archaic drums
varied in proportions of 2:1, 3:1, 4:1, and even 5:1, but the selection of size
was broadly arbitrary. The drums had to be small enough to facilitate ease in
handling and large enough to diminish the burden of producing the ten or
more drums that would be required to constitute each Ionic column at full
height. Since the double-pteron temples of Eastern Greece in the sixth cen-
tury B.C.E. might require more than 100 columns, this was no small issue. But
since the drums were to become “invisible” upon installation, the overall
proportion of column height to diameter would remain the perceptible focus
for expressing architectural rhythm. The proportion of diameter to height
would be inextricably bound to the intercolumniation, and the forest of col-
umns, along with the walls of the Kernbau, would further express the archi-
tectural rhythms through it’s proportional relations. Like epic poetry, whose
literal meaning is delivered not by syllables but by the larger units—the verses
expressed in cola—architectural meanings are expressed, analogously, in larger
units such as the proportions of column diameter to height, intercolumn
distances, and so on. Just as the poet sought to find a compromise between
the rhythms of the verse and the rhythms of it’s meaning, so also the architect
struggled similarly. The meaning was delivered by a selective management of
spaces, revealed by proportional relations among the central parts: the cella,
Kernbau, peristasis, and temple + altar. These dimensions could be traced to
an earlier decision such as the axis width, and earlier still to the width of the
temple walls. And the basic unit, expressed in ancient ells, feet, and dactyls,
whether “2,” “3,” “4,” “5,” or whatever, might have had a symbolic and pro-
ductive meaning for those communities, for it was fundamental to the whole
sacred architecture.
In hexameter verse, intellectual analysis, then, seems inadequate to ex-
plain the poet’s power; so also for the temple. It is likely that the meaning
of the poem is delivered on more than one level; the long and short syl-
lables—in feet and dactyls—might have contributed aspects that shaped and
colored the literal meaning delivered on the level of the verse. The verse and
Robert Hahn 125
content did not run parallel but rather were combined into a seamless whole,
like the erect column. The architect’s meaning was delivered by a perceptual
gestalt; the power of the divine house was transmitted viscerally, not merely
in the overwhelming dimensions but also by the way the spaces, like the
verses, proceeded by an architectural caesura, also carefully calculated in feet
and dactyls. And those large spaces, punctuated by stone—rounded, square,
and angular—were shaped and colored by the smaller increments of detail
such as the antefix, carved surfaces, and even basic block size. The monumen-
tal temples, like the poet’s verse, move us, each in their own special dance.
The temples first push us back by their enormous width and length, and then
by their towering heights, they overwhelm us by producing a kind of dizzying
reverse vertigo. But at the same time, the double-pteron orchestrates a chorus
of light and dark spaces in the petrified forest, highlights an entryway, and
leads our vision inside to the power that displays itself in this vast array of
forms, while still preserving a basic uniformity and harmony. The poet’s epic
takes us on a similar tour, but by a different route. It takes us through the
forest of human emotions, through the rage and pity, courage and fear, and
joy and sadness of our friends and comrades, until finally, we too are led into
the inner depths of ourselves, the skhn– on which all narratives play out.
Pollitt recently made the point that the influence that the tradition of archi-
tectural treatises exercised on the Canon of Polykleitos has not received much
attention.194 The plausible argument that Anaximander’s prose treatise was
influenced and inspired by early architectural treatises—those of Theodorus
in Samos, and Chersiphron/Metagenes in Ephesos—has been proposed just
recently.195 The plausible argument that architectural treatises may have
influenced the writing of treatises on sculpture is this: Since Theodorus,
whom Vitruvius identifies as the first writer of an architectural prose trea-
tise,196 also was a sculptor;197 Skopas was an architect as well as a sculptor;198
Pythios, the architect of the Mausoleum, also was both a sculptor and the
author of a technical treatise on proportion;199 and Arkesios, in addition to
writing commentaries, also carved the cult image for the temple of Asklepios
at Tralles,200 therefore, “the idea that there may have been a connection be-
tween theories of design in architecture and sculpture is not at all implau-
sible.”201 The indebtedness of Polykleitos and his Canon to techniques in
architectural design, in particular the rules of proportion likely detailed in
those earlier prose treatises, suggests a way that we can hear echoes of written
architectural canons now lost. By focusing on the concept of rules of propor-
tion that we can discern from his works, or from reliably accurate replicas of
126 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
The grid technique that was fundamental to the sculptor’s work of the clas-
sical period also was de rigeur for monumental building in the sixth century
B.C.E. in Didyma. When Von Gerkan set out the structure of the temple of
Apollo in Didyma, his reconstruction illustrated the grid drawing that served
as the architect’s plan.214 Although this is the so-called Younger Didymaion,
the reconstruction indicates the architect’s usual procedure. The grid plan
helps us see the fundamentally rectangular structure of the temple plan and
also the square plinths upon which the columns were set, that were integral
to the planning. It also must be emphasized that the size of the plinth was
determined in an inextricable relation to the selection of column diameter. So
while the grid plan emphasizes correctly the square plinth structures, the
square structures must be grasped also within the context of the circles planned
on top of them, and hence the modular role of column diameter. In the
reconstruction of the plan of the temple of Apollo at Didyma, after Von
Gerkan (see figure 2.15), we can see also how the Kernbau expresses the
proportions of 3:1 (length to width), a change from the archaic plan.
The construction of a temple or of a cult building, such as the East
Building on the Sacred Road connecting Miletus to Didyma, was designed
routinely by means of grid plans, and the relative sizes of the parts were
calculated by appeal to a canon of proportions. But how did the architects
begin the numerical assignments in their designs? And moreover, once they
decided upon their basic numbers, how did they make their grids, that is, to
what sort of formulas did they appeal in order to produce the grid? While the
selection of basic numbers, as Schaber had shown, often expressed symbolic
meanings, the numbers reflected also their basic metrology of foot and ell
measurements, and the breakdown of these units into dactyls. The formulas
that their constructions displayed, therefore, are expressions also of these basic
units of measure. Since an ell was equal to 1.5 feet—and thus 2 ells equaled
3 feet—it is hardly surprising that the numbers 2 and 3 predominate in their
building designs. Starting with a grid plan, constructed by the multiplication
of simple numbers expressed in feet or ells, the open spaces were laid out in
round numbers, and to them additional units were added for the installation
of the walls. Since wall width varied from building to building, sometimes as
a function of the enormous height to which the load-bearing structure had
to reach, the routine formula applied first of all only to the open spaces.
While we might be tempted, at first, to describe the horizontal and open
space as “empty,” to distinguish it from the vertical stone construction that
marks its limits, we should remind ourselves that in sacred architecture the
whole space was understood to be pervaded and occupied—horizontally and
vertically—by a spiritual and cosmic power. In Anaximander’s cosmos, the
Òpeiron is the originating source of all things, from which all qualitative
difference derives and back into which it returns, and it should be acknowl-
edged to occupy also the open space that separates the Earth from the heav-
enly wheels. So, while we can speak analogously of spaces being open, we
should be careful not to describe those sacred spaces as empty. Like the
vertical walls that bound out the architect’s open spaces, so also Anaximander
had to add to his formula a unit when expressing the architectural design of
the cosmos, in order to identify the heavenly wheels, like the architect’s walls,
that marked out the boundaries for his open spaces.
Schneider makes his case for the use of the grid technique by the met-
rological reconstruction of the so-called East Building. He infers the grid
plan from the surviving evidence, as do other excavators, by showing that the
precision with which the architect’s worked could not have been achieved by
a casual and unplanned effort at the site. Schneider, like Von Gerkan and
others, shows that the architect’s technique was to make a grid plan, an
informal sketch with numbers and measures attached. The architect then
would set out the plan with the greatest care by measured cords and stakes.
The numerous scratch marks on the stylobates of large buildings, for example,
have been interpreted to demonstrate the informal sketches with which the
architects worked.215 Thus in the absence of any surviving sketches, the archi-
tect-excavator, having made a careful metrological study, infers the grid plan
that was the prelude to the architect’s monumental task. Figure 2.16 shows
Schneider’s reconstruction of the ground plan of the East Building revealed
132 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
FIGURE 2.16 East Building, Sixth Century B.C.E. Miletus/Didyma: Excavation of the
Ground Plan and Metrological Reconstruction of the Grid Plan
Robert Hahn 133
Thus the numbers 16 and 32 express only the open space between the walls,
not the architectural element that rises vertically from the ground plan. When
we factor in the architectural element, the formula for outer width is “a”
“n” + 2, and so 16 1 + 2 ells = 19 feet; the outer length is 16 2 + 2 ells
= 35 feet. In Anaximander’s cosmos, the distances to heavenly wheels can be
expressed by the same architectural formula, as we shall go on to consider,
and to this formula that expresses the open spaces between the Earth and star
wheel, and between the other heavenly wheels, an additional unit was added
134 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
1 foot = 2 2 2 2 dactyls
The formula for constituting an ell from 24 dactyls can be expressed in one
of two ways:
1 ell = 2 2 2 3 dactyls
or, 1 2 3 4 dactyls
While the number “2” belongs directly to the foot system, the number “3”
belongs more directly to the ell system, as we have considered. It has been
conjectured that while the ground plans of archaic Ionic buildings are calcu-
lated sometimes in feet and sometimes in ells, the elevations of those same
buildings are more usually measured in ells.219 This might suggest that the
parts of human architecture that reached upwards into the heavens would
realize a mixture of foot and ell systems, and emphasize more especially the
number “3.” If so, then perhaps Anaximander’s selection of the number “3” in
his cosmic formula expressed appropriately, that is, within his cultural con-
text, the architecture that reached upward through the heavens themselves.
Before we focus directly on Anaximander’s adaptation of this architec-
tural formula, we turn first to consider the metrological system displayed by
the roof of the East Building. The roof of this building is determined com-
pletely within the ell system. Although the complexity of this design goes
Robert Hahn 135
beyond the scope of our concerns here, some considerations of the basic
features of the roof design and its cosmic significance are appropriate to this
discussion.
Schneider provided an analysis of the roof tiles and the patterns that their
installation exhibits. In this way, Schneider provides an overture to explore
archaic ideals by means of two distinct roads: (i) on the one hand, there are
the architectural formulas, revealed by an analysis of the numbers and pro-
portions that the roof construction displays, and (ii) there is the question of
the symbolic meaning of the roof, the man-made “heaven” that sits atop
sacred buildings. Schneider’s study shows that the roof tile formulas are of a
piece with the design patterns operative throughout the entire building. Be-
fore we consider the metrological analysis of the roof construction, let us
consider first what more can be said about the symbolic meaning of the roof
in general.
There is a long history of cosmic symbolism expressed on the ceilings of
sacred architecture in the ancient Mediterranean and Middle East. The evi-
dence from Egyptian tombs and temples, for example, demonstrates the rou-
tine use of ceilings for the expression of cosmic and heavenly themes. In the
earliest tombs to display so-called pyramid texts, the pyramids of King Unas
and Tete, of the fifth and sixth dynasties of the Old Kingdom, the ceilings
of the burial chambers are decorated with stars, and that tradition is followed
a millennium later in the tombs of eighteenth-dynasty Tuthmosis III, and
nineteenth-dynasty Merneptah. Throughout the New Kingdom are elaborate
cosmic scenes. In the tombs of Ramesses IV, VI, and IX of the twentieth
dynasty, for example, the ceilings are adorned profusely with images of Nut,
the sky goddess, swallowing the sun at night and giving birth to it again in
the morning. These motifs appear on many other ceilings. Consider the ceil-
ing design from the tomb of Ramesses IX, in Figure 2.17.220 Thus long
traditions contributed to Greek architecture in the Archaic period, and we
have much evidence that the ceilings were certainly regarded to have cosmic
and heavenly meaning. I have argued here and elsewhere of the signal impor-
tance of the Egyptian traditions and how they were likely imported.221 Can
the case be made for Greek temples and cult buildings in the Archaic period?
The case for the cosmic and symbolic significance of the roofs of Greek
temples and cult buildings must be made differently. Certainly as the column
had cosmic and symbolic significance,222 so also the roof must have had. As
the columns separated or joined Heaven and Earth, as Homer, Hesiod, and
136 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
FIGURE 2.17 Ramesses IX, Valley of the Kings, Luxor: Ceiling in the Burial Chambers, Twen-
tieth Dynasty, c. 1123 B.C.E.
Robert Hahn 137
FIGURE 2.18 Reconstruction of the Akroterion from the Heraion in Olympia, Late Seventh
or Early Sixth Century B.C.E.
Robert Hahn 139
FIGURE 2.19 Painted Reconstruction of the Akroterion from the Heraion in Olympia, Late
Seventh or Early Sixth-Century B.C.E. (after Yalouris)
of its visual dynamism and on other examples of sun symbols from the period.
He concluded that the akroterion is a depiction of the sun.230 He also pointed
out that variations in coloring on the surface show that it was painted; it
represents the image of a solar disk, fiery, circling, and radiating. Yalouris
140 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
noted as well that the conception of the sun (or moon) as a disc or wheel is
not only common to Greeks of prehistoric times but also to the image on
Achilles’ shield, 231 and he suggested that Presocratic philosophers—
Anaximander is named among them—taught with certainty that the sun is
in the form of a wheel.232 The invocation of Anaximander’s name certainly is
appropriate, since he is credited with the view that the sun is a ring of fire,
shaped like the wheel of a chariot. The wheel is hollow, filled with fire, and
what we identify as the sun is nothing but a hole in that wheel. The fire
bursts through the sun wheel, as the fire can be seen at the nozzle of a
blacksmith’s bellows.233 Consider the painted details of the akroterion in Fig-
ure 2.19 that Yalouris reconstructed from the fragments he found at the
archaic temple to Hera in Olympia.234
Yalouris’ suggestion is tempting in another way. Scholars have tended to
give credit to Anaximander for making a model of his geometrically modeled
cosmos,235 and if true, perhaps Anaximander made some sort of model out of
clay. He might have constructed it in concentric, ring-shaped sections, in a
manner similar to the construction of the akroterion. Furthermore, we have two
reports that connect Anaximander to Sparta—one that he had set up a seasonal
sundial there,236 and another suggesting that he offered some advice to the
Spartans about preparing for earthquakes.237 The two different reports connect-
ing Anaximander to Sparta make his visit there seem all but certain.238 Perhaps
in Sparta he saw akroteria of such designs, saw the firing technique for making
them, and in this way, perhaps he made his own cosmic model. Once we have
accepted the idea that Anaximander’s cosmic imagination was fueled significantly
by an architectural inspiration, the archaic temples and cult buildings all offer
the promise of shedding light on his rationalizing cosmology.
If we think through Anaximander’s cosmic model and then review the
architect’s work, we might come to see the akroterion, as the symbol not
merely of the sun but rather of the whole cosmos, in microcosm. Even in
antiquity, Achilles’ shield, which Yalouris finds a suggestive parallel to the
akroterion, was described as a “representation of the world.”239 The sun, then,
might be seen as the outermost ring of the akroterion, radiating its energy in
piercing rays. Atop the Heraion, a temple dedicated to a goddess of fertility,
is perhaps an emblem of the cosmos itself, and not just a symbol of creativity
but a representative embodiment of the cosmic energy that makes organic
growth possible. Anaximander, describing the formation of the cosmos, be-
gins with a surrounding fire, hot and dry, and the appearance of the cold and
moist earth in the center shaped like a column drum. Through a continual
process of change, the fire was somehow separated out into three concentric
rings.240 The process of the evaporation of water from the cold and moist
earth produced mist—aer—that surrounds the rings of fire and accounts for
the fact that the fire is invisible, except for where it shines through the
Robert Hahn 141
apertures. In the image of the akroterion from the Heraion of Olympia, the
outside edge preserves the image of Anaximander’s ring of fire, and the in-
ternal concentric rings, although too numerous to be an accurate depiction of
his cosmos, could be modified simply to produce a faithful likeness. The
visual dynamism of the akroterion design captures the “eternity of motion”
that characterizes Anaximander’s cosmos,241 perhaps even the effect of the
whirling vortex. I have argued elsewhere that in the image of column drum
anathyrôsis, Anaximander’s whole cosmos appears, suggestively, in plan.242 If
the akroterion is reviewed from the same aerial view, it too offers an image
highly suggestive of Anaximander’s cosmos, as a horizontal cross-section
through the plane of the earth, and it shares with it the approximate dimen-
sions of lower column diameter, that is, the module in terms of which the rest
of the building—or cosmos—is calculated.
Therefore, the roof of a Greek temple took on cosmic and symbolic
meaning as the sacred heaven enclosing the holy, divine Kernbau, the space
of sekos and adyton.243 The roof of a cult building expressed also such cosmic
meanings when it was adorned by an akroterion. In the akroterion, certainly,
a cosmic sense was transmitted immediately to all who came into its view.
Perhaps the immediate impact was the image of the vibrant sun, but in any
case a heavenly symbol. Achilles’ shield depicted the world in plan view, and
Anaximander’s map of the oÎkoum°nh and his cosmic map were imagined
also in plan view. The akroterion can be viewed as a plan, but once installed,
the plan appears as an elevation. The architectural ground plan and the el-
evation of the temple, as Schaber had shown, display also this interrelation.
Might it be, then, that we can get yet another architectural hint of how
Anaximander imagined the projection of his cosmic plan into the elevation
that the three-dimensional cosmos must be? Could it be that an analysis of
the patterns of the roof tiles reveals, or confirms further, the architectural
formulas according to which the building was constructed, or at least echoes
the formulas that found expression in the ground plan and in the elevation?
And if Anaximander thought through the structure of the heavens in terms
and formulas by which the architects produced man-made heavens, might he
have reckoned analogously his own cosmic architecture?
Schneider studied some 6000 fragments of the clay roof from the East
Building. He concluded that the entire design of this archaic building and its
roof, right down to the details of the roof ornamentation and coloring, fol-
lowed a logically coherent metrological system. That system is derived from
the basic divisions of the Ionic units of measurement and consists in foot and
ell, and from their relation of 16 to 24 dactyls (2:3), as we have already
considered. The roof construction is based completely on the ell system and
based on a quadratic ell plan. In Figure 2.20 we can see how the real dimen-
sions of the roof tiles were established by adding or subtracting dactyls (D)
142 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
FIGURE 2.20 Metrological Reconstruction of the Plan for the Flat Tiles of the Roof, East
Building, Miletus/Didyma, Sixth Century B.C.E. (after Schneider)
from the same formula used to design the ground plan. Between each flat tile
was a space, part of the metrological plan, and over that space a cover tile was
placed. When the cover tile was installed, part of the flat tile is overlapped,
as can be seen in Figure 2.20. The roof formula applies, first of all, to the
open space created by the size of the tile itself, and then to the open spaces
constituted by them collectively. The formula, again, is “a” “n.” The flat tile
width can be expressed by the formula “a” “n”–2, where “a” equals 24
dactyls and “n” equals 1, thus 24 1 – 2 = 22 D; and the flat tile length is
expressed by the formula “a” “n” + 1, where “a” equals 24 and “n” equals 1,
thus 24 1 + 1 = 25D. In Figure 2.20, the flat tile is shown, and the
additional space between each flat tile where the cover tile would be placed
is also indicated.
Schneider’s reconstruction shows how the whole form of the tiled roof
can be expressed by the same architectural formula by which the architect
designed the ground plan and the elevation. The reconstruction showed that
24 rows of flat tiles were needed to cover the length of the roof of the East
Building, and 16 rows of flat tiles were needed to cover the width. From this
discovery, Schneider suggested that a rhythmic, three-ordered system was
employed, in the proportions of 2:3.244 In the roof system of an ell design,
Robert Hahn 143
every tile reduplicates itself in the whole form of the roof. Each is a micro-
cosm of the macrocosm represented symbolically as the roof. So, while each
flat tile is ±24 dactyls in width and length, each row consists also of 24 tiles,
that is, 24 ells. This base number is depicted within the 24 braid elements of
the colored relief of the eave’s edge. And because the rands of the roof are
built by special sima tiles, the number of regular flat tiles in a horizontal row
also can be expressed by the same formula “a” “n.” In this case, the formula
for the flat tiles in a horizontal row can be expressed so that “a” equals 24 tiles
and “n” equals 1, thus 24 tiles 1 – 2 sima tiles = 22 flat tiles. The same
formula also can be shown to express the design of the vertical rows, but in
a more complicated way since the sizes of the tiles are not uniform but
diminish from the eaves to the ridge. The reconstruction of the roof of the
East Building, after Schneider, is presented in Figure 2.21.245
FIGURE 2.21 Reconstruction of the Flat Tiles and Cover Tiles of the Roof of the East
Building, Miletus/Didyma, Sixth Century B.C.E. (after Schneider)
144 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
Anaximander seems to have adopted, the distances to the heavenly wheels are
reckoned by adding a unit—“1”—to each calculation. If Anaximander had
adopted the architectural formula, the structure of the cosmos could be ex-
pressed as “a” “n” + 1, where “a” is a constant and “n” is a variable expressed
in a series of progressions, and the “+ 1” represents the concrete world body—
the architectural wall, as it were—the earth diameter, which is 1 unit thick.251
Starting with the number 3, Anaximander’s cosmic distances can be
expressed in terms of this formula. The value of “a” is 3 3, and “n” is
expressed in a series of proportional progressions—3 1, 3 2, 3 3—
in a manner comparable to the formula by which the architects designed the
roof, the man-made heaven. The addition of “+ 1” is contributed to each
formulation and represents the concrete world body, the diameter of the
wheel. Thus Anaximander’s cosmic distances could be expressed as follows,
and represented in Figure 2.22.
Note, that in all of these series, the formula is that the module number, “3,”
is multiplied by itself, and then again is multiplied in the progression of the
sizes—x 1, 2, and 3—and each pattern is doubled. In the standard
formula, “a” is 3 3 and “n” is a progression 3 1, 3 2, and 3 3, where
the initial “3” is an approximation of to express the circumference of the
wheel. The result in each case is then doubled, to express the diameter, not
radii. This formulation harkens back to Schaber’s description of the principle
of doubling that seemed to have been fundamental to the cosmic fertility and
fecundity that the archaic architects sought to express. Thus in the pattern of
organic growth, the radii of the wheels grow in size by increments of “3
3” or 9 units—9, 9 + 9, 9 + 9 + 9—and the sizes of the wheels grow more
Robert Hahn 147
FIGURE 2.23 Anaximander’s Architectural Formula for the Cosmos: The Sizes of the Cosmic
Wheels
148 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
or less in increments of 50 modular units, just like the usual sequences of the
archaic temples—axis width, cella, Kernbau, (and peristasis). The sequences for
the archaic Ephesian Artemision D, 44, 88, and 132 (ells), and for the archaic
Samian Heraion Dipteros II, 48, 107, and 155 are surprisingly comparable to
the threefold sizes of Anaximander’s heavenly Kernbau—54, 108, and 162.
The temple Kernbau, as we have considered, is the sacred enclosure. It
is sometimes referred to as the Cella Building, or even the Sekos, depending
on whether or not it has a roof. It is tempting to think that Anaximander
imagined the limited or defined cosmic space—defined in outermost dimen-
sions by the wheel of the sun—as a cosmic Kernbau. It shares comparable,
proportional, sacred dimensions and reflects a comparable design technique.
This means that if we have the numerical assignments identified correctly, the
size of the cosmos grows arithmetically—by additions of 54 units—but the
augmenting distances in the cosmos grow in a geometrical series. Like the
growth of the Ephesian Artemision, where the size of the temple can be
reckoned in increases of 44 units, grounded in the number 4, or the Heraion,
where additions are based on increases of 50, grounded in the number 4,
Anaximander’s cosmos grows in increases of 54, grounded in the number 3.
The argument is not that Anaximander copied the architects any more than
the latter copied each other. Rather, what they shared in common was a
certain formulaic method—like the poets and sculptors—testified and reified
in Ionian architecture. Despite the differences in numerical assignments,
Anaximander’s cosmic architecture is clearly comparable to the terrestrial ar-
chitecture that sought to express the cosmic and divine powers created by the
architects. The reckoning of Anaximander’s cosmic architecture is based on
the module 3:1 (or 1:3), and hence expressible in design patterns, like the
holy Kernbau, that the architects created contemporaneously.
The plausibility of this thesis and the analogy upon which it rests depend
upon the understanding that Anaximander’s cosmic imagination was inspired
by the architects and the monumental, cosmic houses that they were creating.
Those projects brought together ancient religion and rationality, oxymoronic
as they first might seem, in a way that had no precedent in Anaximander’s
world; side by side, while the temple testified to the meagerness of mortal
wisdom in the presence of the inscrutable divine reality, it was, in fact, erected
by a rational mentality that placed a premium on rational, not mythological,
accounts. The rationalizing architects, by their awesome productions, and
probably through discussions at the building sites, urged Anaximander to
think, on their analogy, of the cosmos itself as a great house. Should we be
surprised then to discover that the proportions that seemed symbolically
appropriate to an archaic audience were called upon to construct the divine
and cosmic houses—the temples and other sacred architecture? And further-
more, should we be surprised to see that Anaximander called upon just these
Robert Hahn 149
same formulas when he offered his original contribution to such cosmic dis-
cussions, at just the time and in just the place where monumental temple
architecture transfixed the communities of Eastern Greece?
EPILOGUE
40. For example, since Tannery (1887/1930, 94ff ), followed by Diels (1897),
Burnet (1920), and Heath (1921). Cf. also Jaeger (1939), p. 157, and Kahn (1960),
who accepts it provisionally as a good guess, p. 88, and also G.E.R.Lloyd (1996, 174).
41. For the sun wheel, the number 27 is offered both by Hippolytus (Ref. I, 6,
4–5) and Aetius (II, 21, 1). However, Aetius (II, 20, 1) also gives the number 28 for
the sun wheel.
42. The only numerical assignment for the moon wheel comes from Aetius (II,
25. 1).
43. Aetius (II, 15, 6).
44. Cf. Kahn (1960, 77ff ).
45. Cf. Lloyd (1966, on “analogy”).
46. Cf. Kahn (1960, 62); Kirk-Raven (1971, 136ff ). The most recent survey is
provided by Naddaf (1998), who believes, mistakenly, that Anaximander meant the
numbers to refer to diameters, not radii, and so sizes, not distances.
47. Hahn (2001, ch. 4).
48. This point has been made elegantly by Couprie (1998).
49. Cf. Kemp and Rose (1991). Although they are addressing ancient Egyptian
design techniques, specifically the theories promulgated by Badawy (1965), the argu-
ments apply ceteris paribus.
50. Cf. Schaber (1982), whose metrology of both the seventh century and sixth
century Artemisian “C” and “D” suggest that central to them were numerical assign-
ments we familiarly call “Pythagorean.” Central were the numbers 1, 2, 3, 4, and 10,
according to his metrological studies.
51. Detailed in Hahn (2001, ch. 4).
52. On the authority of Agathemerus I, 1–2. Discussed profitably by Jacoby
(1912) in RE, VIII s.v. Hekataios (cols. 2667ff; 2702–7), and Heidel (1937, 11ff ).
53. Cf. the tripartite rendition in Robinson (1968, 32).
54. For the architectural techniques of scratch marks and the supposition that
there must have been informal drawings that guided the markings, cf. Kienast (1985,
esp. pp. 111–112). Reuther (1957) already drew attention to the elaborate system of
scratch marks. The leader in these arguments was Petronotis (1972). Cf. also the
review of Petronotis by Wesenberg (1976).
55. Kahn (1960, 83n3).
56. Herodotus (iv, 36), ™p¿ t¬rnou, by either a compass or a lathe.
57. The dimensions of 3 1 are attested to by Ps.-Plutarch, and echoed by
Hippolytus (Ref. I, 6, 3).
58. On the authority of Aetius (II, 21–1).
59. Cf. the details in Hahn (2001, ch. 4) tracing the number “9” in Homer and
Hesiod.
Robert Hahn 155
60. Cf. the examples in Homer’s Odyssey 10, 19; 10, 390; 14, 247.
61. Cf. Hesiod Theogony 722–726 for cosmic distances, and elsewhere such as
790 for the usual formula where “9” suggests great multitude.
62. This is precisely how Couprie (1998) puts it.
63. Homer (Iliad 2, 327ff ).
64. Homer (Odyssey 14, 240ff ).
65. For examples, cf. Homer (Iliad 1, 52ff ); and Odyssey (9, 82ff; 10, 27ff; 14,
228ff ).
66. Gigon (1945, p. 91).
67. Hesiod (Theogony 790); cf. also the Hymn to Demeter 47–51.
68. Hesiod (Theogony 722–726). Cf. also the “9+1” formula at 775–805.
69. This has long been emphasized; cf. Vlastos (1947) and Vernant (1983, p.
185).
70. Krischen (1927, 76, 89).
71. Reuther (1957, 60).
72. Gruben (1963, 94).
73. Wesenberg (1983, 31ff ).
74. Cf. DK (12A10.33); Psuedo-Plutarch, (Strom. 2).
75. Homer (Iliad 18, 483ff ); Cf. Hahn (2001, ch.4).
76. Tannery (1930, 91); Diels (1923, 65–76, esp. 72).
77. Usener (1903).
78. Ibid. 1–47.
79. Ibid. 161–208.
80. Ibid. 321–362.
81. Ibid. 348.
82. Ibid. 360.
83. Von den Steinen (1897, 84ff ).
84. Locke (Essay, Bk. II, ch. xvi, sect. 6). Also, Locke refers to a tribe in Brazil;
tribe members had no names for numbers above the number “5” and thus used their
fingers and those of others who were present to express any number beyond that.
85. Plato (Philebus 16Dff ); cf. also Hahn (1978) on Philebus 15B1–8 for the
background problem.
86. Cf. Hahn (1995, 97–98).
87. Cf. Hahn (2001, ch.3).
156 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
88. Cf. Orlandos (1965, II, 100); Martin (1968), Coulton (1977, 46–48).
89. Cf. Hahn (1992).
90. Cf. Aristotle. (de Caelo 295b10–14).
91. Cf. Hahn (1995, 118–119).
92. On the arguments by analogy, cf. G.E.R. Lloyd (1966).
93. This act of “un-concealing” literally was telling the truth, ™-l–qeia. Cf.
Heidegger (1975).
94. Cf. Hahn (1995, 97–98 and 2001, ch.2), tracing the treatment of the column
in Homer, Hesiod, Pindar, and Plato.
95. Cf. Dinsmoor (1955).
96. Euclid (I. 2).
97. The illustrative diagrams are by Nils Hellner, Technical University of Munich,
and are included here with his kind permission.
98. Permission to publish these sketches was kindly given by the author, Nils
Hellner, of the Technical University of Munich.
99. Illustration courtesy of Nils Hellner. There also is evidence that as the Dipteros
II construction continued later on, a four-part marking system was adopted, although the
three-part one was adequate. The later technique involved four markings at the end of a
rectangular cross, with its center coinciding with the center of the circle; this marking
system was founded upon the division of the circle into four regions of 90° each.
100. Kienast (1986/87); cf. the section on “The Planning and Construction of
the Tunnel.”
101. For example, cf. Toulmin (1960) who supposed not only that the hole could
have allowed for fresh air but also supposed that the construction was made possible
by means of a method of triangulation, literally similar triangles. Toulmin’s hypoth-
eses, daring as they were in 1960, have been discounted largely by Kienast’s (1986/87
and 1995) studies.
102. This is central to the thesis of Schaber (1982).
103. Schaber (1982).
104. Baldry (1932, 27–34).
105. Cf. Hahn (1995 and 2001, ch. 2). Cf. also Schibli (1990) on Pherecydes, esp.
128ff.
106. Cf. Hahn (2001, ch. 4).
107. Early temple architecture relied on trees to hold up the roof. As the temple
became significantly enlarged, the trees were no longer capable of handling the heavier
loads.
108. Kahn (1960, Appendix I, 219–230).
Robert Hahn 157
147. For example, perhaps the wall width had to be at least 3 feet to support the
entablature and roof. Then the choice of any measure greater than the structural
requirement—such as 4 feet in the Ephesian Artemision “C” and Dipteros II in
Samos—might plausibly have been for symbolic reasons. Of course, one reason some
of the archaic buildings seemed heavy might have been the result of ensuring sufficient
structural stability, and due to inexperience, the architects chose to err on the side of
supplying greater support, even if unnecessary. But the likelihood that the archaic
Ionic columns were 9 or 10 times the lower column diameter, unlike the archaic Doric
columns that stand in proportions roughly 5 or 6 times the lower diameter, would
seem to count against such a generalization.
148. Cf. Wesenberg (1983, 25); “round numbers” is an expression that should
include fractions such as the following: 1/2, 1/3, 1/4, 3/5, 9/10.
149. Smith (1925, vol II, 214). On fractions in general, cf. Heath (1921).
150. Coulton (1975, 98).
151. Coulton (1975, 85).
152. Coulton (1975, 99).
153. Vitruvius III.3.7. Cf. also Wesenberg (1994, 92): Modulus ist in der ionischen
Version der untere Säulendurchmesser.
154. Krischen (1927, H. 10–12) [Hochbauteil], 76, 89; ders. 1923, 6ff.; 1956, 68.
155. Reuther (1957).
156. Gruben (1963).
157. Bammer (1984).
158. Wesenberg (1983, 21): ”ud” = “unterer Säulendurchmesser, gemessen oberhalb
des Ablaufs.”
159. I am grateful to Nils Hellner for suggestions as I prepared Figure 2.8, and
I am grateful to Burkhardt Wesenberg for suggestions as I prepared Figure 2.9. For
more complex drawings of the Ionic and Doric order, see Ginouves (1998, plate 36),
and Lawrence/Tomlinson (1996).
160. Cf. Wesenberg, (1983, 23). Thus the architrave is exactly the height of the
fitted section, namely, geison and sima combined.
161. “SH” = Säulenhöhe.
162. “GH” = Gebalkhöhe.
163. “AH” = Architravhöhe.
164. Wesenberg (1983, 26ff.).
165. Ibid. 29; Vitruvius also suggests that the proportions 1:8, 1:8.5, 1:9.5, and
1:10 may all be applicable to Ionic columns, and thus 1:9 is already a historical
proportion in his view.
166. Ibid. 31.
160 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
167. Cf. Coulton (1977, 67); cf. also Wesenberg (1983, 25).
168. Since the column bases contained inscriptions proving that they were dedi-
cated by Kroisos, and since Kroisos came to power c. 560 and perishes at the hands
of Cyrus in 546, there is good reason to suppose that by his patronage, significant
building took place following his rise to power. Cf. also Hahn (2001, ch. 2).
169. Schaber (1982, 106, 110).
170. Schaber (1982, 115).
171. Schaber (1982, 117).
172. We have already noted the importance of “10” in the formula “9 + 1” in
archaic poets such as Homer and Hesiod.
173. Schaber (1982, 116).
174. Schaber (1982, 119). Naturally, the determination of the exact size of the
foot or ell is crucial to the metrology, and Schaber devotes a considerable part of his
book to this debate among excavators over the proper assignments. The ratio of
foot:ell = 3:2 is not in doubt. Cf. also Nylander (1970, 97), and Schneider (1996, 34).
175. Indeed, all the numbers in Schaber’s reconstruction of the elevation are
conjectural. They are based upon an appeal to the rules of proportion alone, not the
preserved artifacts, since none survive in situ.
176. It must be made clear that there is no surviving evidence in situ to confirm
the reconstruction of the elevation. Schaber must rely on the surviving evidence of the
ground plan and an appeal to the rules of proportions to make the case for his
reconstruction.
177. Schaber (1982, 83).
178. Ibid., 82.
179. Ibid., 84, makes the point that the parts of the column stand in relationship
to one another, and to some “perfect” number.
180. Cf. Coulton’s introduction of this English expression “Cella Building” in
Ginouves (vol. 3, 39), to describe the Kernbau, the structure inside of the colonnade
of a temple (i.e., any combination of cella (sekos) with pronaos, adyton or
opisthodomos).
181. Cf. Schwaller de Lubics (1957, esp. “Mathematical Thought,” 88–189) for
an exposition of the formative meaning of numbers for the Egyptian architects.
182. Schaber (1982, 116).
183. Ibid., 119.
184. Historians of philosophy should pay careful attention to this idea in archi-
tectural design when reflecting on Aristotle’s attribution to Plato of the doctrine of
Ideal Numbers. In that discussion, Aristotle explains that this was Plato’s central
doctrine, the generation of the Ideal Numbers from two principles, “The One” and
the “Indefinite Dyad.” The Indefinite Dyad is a principle of doubling and halving.
Robert Hahn 161
185. Cf. Herodotus (iv.62). Cf. also LSJ, 1592, “sacred enclosure.”
186. Gruben (1963, Abb.1), but see also 83, 84, 90–91.
187. Fehr (1971/72, 14ff, and 1998, 163).
188. The metrology is derived from Schleif (1933), and Buschor (1930). The
measurements are conjectural, based upon reconstructions, and when the reconstruc-
tions are so seriously in doubt, no numbers are supplied.
189. The metrology comes from Nils Hellner, of the Technical University of Munich,
in recent correspondences. His numbers differ slightly from those obtainable from Reuther
(1957). Hellner presumed that the ell measured .524 meters, because the front facade
of the temple measures 52.45 meters, and the temple plan has been assumed to be 100
ells in width. Schneider (1996, 34), supposed that the ell measure in Didyma was .5235
meters, that is, for all practical purposes it was identical. Schaber (1982, 89) reached the
conclusion that the Kroisos ell was .5236 meters, and this unit generally is regarded to
be identical with the Egyptian ell; cf. also Petrie (1883), and Cole (1926).
190. Derived from Gruben (1963), calculated from Abb I, but also cf. pp. 83, 84,
90–91.
191. Cf. Hahn (2001, ch.3).
192. DK 22B90.
193. DK 22B49a.
194. Pollitt (1995, 20). Sources apply the word “Canon” both to a statue by
Polykleitos, usually the Doryphoros, and to the written theory embodied by his works.
Galen’s de Placitis Hippocratis et Platonis 5 is our source for the written treatise.
195. Hahn (2001, ch. 2), and outlined earlier in this monograph.
196. Vitruvius VII.12.
197. Oberbeck (1868, 1350–63); more recently, see Holloway (1969).
198. Pausanius 8.45.4.
199. Vitruvius VII.12–13, also I.12; Cf. also Pliny NH 36.31.
200. Vitruvius 7.11.
201. Pollitt (1995, 20). There also are other artists, whom Pollitt does not men-
tion here, about whom we have reports that they, too, authored treatises, including
Agatharcos, Euphranor, and Apelles. The reports come from Vitruvius 7.11, and Pliny
NH 35.129; 35.80.107.
202. Diodorus Sicilus 1.98.5–9, discussed in Burton (1972, 284–290). Cf. also
Holloway (1969) for the unconvincing conjecture that this “Theodorus” is not the
same one named in conjunction with Rhoikos.
203. Cf. Hahn (2001, ch.2).
204. Iverson (1975, 75–88).
205. Cf. Robins (1994, 160ff.).
162 PROPORTIONS AND NUMBERS IN ANAXIMANDER AND EARLY GREEK THOUGHT
Dirk L. Couprie
PROLOGUE
Adumbrare necesse erat astronomiae graecae initia, quae quanti momenti
essent, sero intellegi coeptum est.
—H. Diels, Doxographi Graeci
Many craters on the moon bear the names of famous philosophers and as-
tronomers. A modest crater at 66N, 48W, which is not even indicated on
many maps of the moon, has been named after Anaximander. This part of the
book on Anaximander’s astronomy is written with the conviction that he
deserves more. Actually, I am convinced that Anaximander of Miletus (610–
546 B.C.E.) was one of the greatest minds in history. I do not hesitate to put
him on a par with someone like Newton. This sounds even more astonishing,
since so little of Anaximander’s thoughts has been passed down to us: one
sentence, a few loose words, and a handful of reports, written by Aristotle and
doxographers, some of whom lived ages later. In many surveys of the history
of philosophy, Anaximander is known for having introduced “the Boundless”
(t¿ Òpeiron) as the principle (™r‹) of all things, and as the author of the
first philosophical line that has been preserved. We do not know, however,
what exactly he had in mind with the Boundless. Did he mean something
without spatial or temporal limits, or both, or did he mean something quali-
tatively indefinite? Many authors have speculated about questions such as
these. One-half of the scholars who studied his one remaining sentence ar-
gues that it refers to the Boundless, whereas the other half maintains that it
does not. Relatively many testimonies of the doxographers, approximately
one-third of them, have to do with astronomical and cosmological questions.
These testimonies contain—sometimes strange—images, like chariot wheels,
a column drum, the bark of a tree, and lightning fire. Whatever the link has
been between all of these images and the Boundless, or how they relate to the
one remaining sentence, can only be guessed. For reasons such as these, I
think that there is no use in trying to give a coherent, overall exposition of
Anaximander’s teachings. What I present here is, therefore, nothing more
than a survey of his astronomical ideas, the essence of which I prefer to label
his “discovery of space.”
One of the concepts most characteristic of Western thought, and distinc-
tive for Western civilization, is the concept of space—the space of the uni-
verse. Western man is the conqueror of space. Western civilization has sent
people from the earth to the moon and has sent spacecraft beyond the far-
thest planet of our solar system. We search for signs of extraterrestrial civi-
lization in our galaxy, and we peer through telescopes billions of light-years
into space. Whether one regards it as a blessing or not, the fact is that no
other civilization has ever accomplished such astonishing feats. These are
167
168 THE DISCOVERY OF SPACE
triumphs of Western technology. But they are not only technological achieve-
ments. In order to send spacecraft to other planets, one must be convinced
of the possibility of doing such things. I am not talking here about the
technological capability but about the conceptual possibility. All of the achieve-
ments mentioned presuppose a particular concept of the universe, which may
be called the Western paradigm of the universe. When one conceives of the
heaven as a firmament, a big dome or ceiling, or an inverted bowl, onto which
the celestial bodies are glued in some way or another, one never will even
consider the possibility of firing people off in rockets in order to be able to
walk on other planets. In other words, the Western concept of the heavens
is that of a, so to speak, three-dimensional universe, a universe with depth,
in which the celestial bodies lie behind one another. One has become so used
to this concept that it is difficult to recognize how unique and, in a sense,
how unusual it is. For example, if you have ever been in a planetarium, where
lights, projected on a semispherical ceiling, represent stars, planets, the sun,
and the moon, perhaps you may realize that this is the way we see the uni-
verse. The depth of the heavens is not something we see but something we
know. We do not see that the celestial bodies are behind one another, but we
know that this is the case. The universe, conceived of as three-dimensional,
with depth, is a highly artificial one, and it is completely different from the
more natural way of looking at it, as a firmament onto which the celestial
bodies are glued, all at the same distance.
How did this idea of space, this unique concept of a stratified universe,
originate? I argue that it was the discovery of a Greek, who lived in the Asian
part of Greece in Miletus in the first half of the sixth century B.C.E., and
whose name was Anaximander. He taught three epoch-making theories: (1)
that the celestial bodies make full circles, and thus go underneath the earth;
(2) that the earth floats free and unsupported in the center of the universe;
and (3) that the celestial bodies lie behind one another: first the stars, then
the moon, and farthest away the sun, at nine, eighteen, and twenty-seven
earth diameters’ distance from us. The important thing in this last theory is
not that he was wrong about the order of the celestial bodies, nor that his
estimates of their distances were far beyond the truth, but that he said that
they were behind one another. In the following pages, it will be argued that
these three ideas together, and the third one in particular, mark the origin of
the Western world picture. Anaximander was the discoverer of space.1
One gets an impression of the enormous difference between Anaximander’s
idea of the universe and older mythological ones, when comparing it to the
Egyptian representation of the heaven goddess, Nut, held up by the sky god,
Shu, who stands upon the earth god, Geb, as shown in Figure 3.1.2 This
picture, many specimens of which exist, has barely changed over the centu-
ries. I return extensively to this subject in a later section of this part. Another
Dirk Couprie 169
FIGURE 3.1 Egyptian Painting of the Heaven Goddess, Nut, Held up by the Sky God,
Shu, Who Stands upon the Earth God, Geb
FIGURE 3.2 The Covering of the Holy Tree with the Mantle of the Heavens on an
Assyrian Seal Cylinder
see the light of the sun and of the moon like a stream of lightning fire.8
Sometimes the opening of the sun’s or moon’s wheel closes: then we experi-
ence an eclipse. The aperture of the moon wheel also gradually opens or
closes when the moon is crescent or waning. The doxography gives us some
figures about the dimensions of Anaximander’s universe: the sun wheel is
twenty-seven or twenty-eight times the earth, and the moon wheel is nine-
teen times the earth. The wheels of the sun and moon are said to lie aslant.
It is related that Anaximander knew about the turnings (tropa¥) of the sun
(viz., the solstices) and the moon. These turnings have meteorological causes,
namely, winds, which in their turn arise from the drying of the original
moisture by the heat of the sun. Anaximander’s universe has come into being
172 THE DISCOVERY OF SPACE
“when a germ, pregnant with hot and cold, was separated [or separated itself ]
off from the eternal, whereupon out of this germ a sphere of fire grew around
the vapor that surrounds the earth, like a bark round a tree” (Diels and Kranz
[1951–1952], 12A10).
This is what the doxography tells us about Anaximander’s astronomical
ideas. At first sight we recognize them as being rather primitive and obscure,
pre-Ptolemaic astronomy. Some authors even think that they are so bizarre
and confused that it is better to give up trying to offer a satisfying, coherent
interpretation.9 My aim in the following pages is to show that Anaximander’s
astronomy, strange as it seems to be, makes perfect sense, and that it stands
at the cradle of our Western way of understanding the universe. Therefore, if
we attempt to understand Anaximander, we may understand ourselves some-
what better. On the other hand, in interpreting Anaximander’s thoughts, a
characteristic difficulty arises. It is the difficulty to identify oneself with some-
one who thinks that the earth is flat, that it is located at the center of the
universe, someone who has no other astronomical instruments at his disposal
than the naked eye and a gnomon. In order to overcome this difficulty, we
have to perform some mental gymnastics. Time and again we shall see how
scholars, from the earliest doxographers up to our times, have been misled by
what I call “the anachronistic fallacy.” To become accustomed to “looking
through Anaximander’s eyes,” we begin with some exercises in ancient Greek
astronomy. First we shall pay attention to the different forms of the anach-
ronistic fallacy, then we will do three exercises in looking at the heavens with
the naked eye; subsequently, we shall learn what kind of observations can be
made with the help of a gnomon, and in addition we shall try to show the
astronomical implications of a reconstruction of Anaximander’s map of the
world. Having laid this base, we will be in a position that enables us to
understand in what sense Anaximander’s astronomy may be regarded as the
paradigm of the “Western” way of looking at the universe. I will explain
Anaximander’s discovery of space mainly by interpreting the numbers he used
in denoting the distances from the heavenly bodies to the earth. After this,
I will try to visualize Anaximander’s conception of the universe. In a separate
section, the hypothesis will be put forward that some puzzling pictures of the
heavens on Ptolemaic Egyptian temple ceilings can be seen as influenced by
Anaximander’s theories. In a concluding paragraph, something more will be
said on Anaximander’s influence.
All of this is done with the conviction that the first and most important
way of understanding Anaximander’s astronomical ideas is to take them se-
riously and to treat them as such, that is, as astronomical ideas. In other
words, if Anaximander’s astronomy makes any sense at all, it must be possible
to look at the heavens and recognize what he meant. It will appear that many
of the features of his universe that look so strange at first sight make perfect
Dirk Couprie 173
The doxography has two conflicting stories about what the earth was like in
Anaximander’s view. Diogenes Laertius tells us that Anaximander taught that
the earth was spherical.10 Pseudo-Plutarch, Hippolytus, and Aetius, on the
contrary, said that, according to Anaximander, the earth was cylindrical, like
a column of stone.11 Generally, modern scholars agree that Diogenes Laertius
was mistaken. They also agree that the words “column of stone” must mean
“column drum,” as suggested by the ratio 3:1 for the diameter and the height,
given by Hippolytus.12 The main reasons given are that Diogenes Laertius’
reports are not always reliable, that the image and the words into which it is
put in the other reports are so curious and yet precise that they must be
authentic, and that the cylindrical form is a kind of missing link between the
flat earth of the primitive world picture and the spherical earth of the more
sophisticated concept from Plato and Aristotle onward. Therefore, we may
say that Anaximander certainly did not teach that the earth was spherical.
Some scholars have wondered why Anaximander chose this strange shape.
The strangeness disappears, however, when we realize that Anaximander
thought that the earth was flat and circular, as suggested by the horizon. Such
was the idea of the shape of the earth at that time, as proven by Homer’s
description of Achilles’ shield,13 and by the contemporary Babylonian map of
the world (see the reproduction in Figure 3.17). For one who thinks, as
Anaximander did, that the earth floats unsupported in the center of the
universe, the cylinder shape lies at hand.
Why is it so important to stress that Diogenes Laertius was wrong,
although it is no point of discussion among scholars? The cylindrical earth is
a kind of permanent reminder of what I will call the danger of the anachro-
nistic fallacy. This fallacy results from our inability to give up our familiar
interpretation of the universe, and to look upon Anaximander’s astronomical
174 THE DISCOVERY OF SPACE
teachings with an unbiased eye. We shall meet this fallacy over and over
again, and in various disguises. Actually, Diogenes Laertius is a good example
of an early victim of the fallacy: he simply was not able to imagine that
anyone could believe the earth to be something other than a sphere.
Since Aristotle’s time, the earth as well as the universe as a whole, has
been conceived of as a sphere or, better, as a number of concentric sphere
shells. This image of a spherical universe has been strengthened by the con-
cept of the sphere as the perfect body. The image is so strong that it is hard
to imagine that people before Aristotle might have had another conception
of the universe. Almost all commentators take it for granted that Anaximander
too taught that the universe was spherical. But did he really? We read that
the earth is cylindrical, that the celestial bodies are wheels, like those of a
chariot,14 and that during the genesis of the universe a sphere of fire grows
around the air that surrounds the earth “like the bark around a tree.”15 The
common view is that this image of the bark around a tree must go back to
Anaximander himself.16 If we try to look at the doxography with an unbiased
eye, it is not so much the image of spheres that pops up, as that of cylinders
and wheels. The very word “sphere,” used here by Theophrastus in DK 12A10,
has to be viewed as an anachronism: even when describing Anaximander’s
image of telescoping cylinders (the bark around a tree), Theophrastus is not
able to abandon the spherical universe of his teacher, Aristotle. I agree fully
with Furley, that “it should be noticed that the idea of circular orbits for all
the heavenly bodies does not itself entail that the whole cosmos is spheri-
cal.”17 The same point has been made already by Heidel, who explicitly calls
it an anachronism to ascribe to Anaximander the idea of a spherical cosmos:
“There is in fact nothing to suggest a spherical cosmos in the scheme of
Anaximander: everything points rather to a system of concentric circles.”18 If
we want to understand Anaximander’s astronomical thoughts, we had better
say that he did not teach the sphericity of the universe.
The difficulty of imagining Anaximander’s cylindrical earth entails an-
other anachronism. Pliny tells us that Anaximander understood the obliquity
of the zodiac, or ecliptic.19 It is a matter of dispute among scholars whether
this tradition is truthful or not. The main arguments for doubting it are that:
(1) even in Babylonian astronomy, the first time the zodiac is mentioned is
in a cuneiform tablet, dated about 410 B.C.E.,20 a considerable time after
Anaximander; (2) the same discovery also is ascribed to Oenopides,21 who
lived a century or so later; and (3) it is not in accordance with Anaximander’s
rather primitive astronomy. This latter point can be made stronger. The zo-
diac, that is, the broad band of constellations through which the planets, sun,
and moon wander, is not the same thing as the ecliptic, which is the sun’s
path through those constellations. The zodiac is presumably a much older
concept, and we may assume that the Babylonians, who paid much attention
Dirk Couprie 175
to the rather complex movements of the planets, were acquainted with it,
even though there seems to be no written evidence for this knowledge. That
the sun describes a circular annual path along those stars is more difficult to
observe, be it alone the fact that the brightness of the sun itself is a big
hindrance to an exact determination of its place among the stars. But most
of all, the obliquity of the ecliptic is a concept that belongs to the doctrine of
a spherical earth within a spherical universe. The plane with respect to which
the ecliptic is inclined is that of the equator of both the spherical earth and
the spherical universe. Anaximander’s earth was cylindrical and, consequently,
the concept of the equator does not make sense in this context. Or, if it does,
“equator” must mean something quite different on a flat earth, as will be
shown in the discussion of Anaximander’s world map. So we may conclude
that Anaximander did not teach the obliquity of the ecliptic.22
The doxography tells us that Anaximander conceived of the heavenly
bodies as rings or wheels.23 The last word seems, according to Diels/Kranz,24
to be the most authentic, and, I might add, also “wheel of a chariot” and “rim
of the wheel.”25 I completely disagree with Kahn, who suggests that, on the
contrary, “it seems more likely that” the image of celestial wheels “reflect(s)
the picturesque style of some Hellenistic popularizer.”26 The word “ring” is,
I think, an anachronistic expression that belongs to later models of the uni-
verse called “armillary spheres.” In the doxography, we still can see how later
authors wrestled with Anaximander’s concept of wheels, which was strange to
them. Achilles Tatius definitely does not understand it any longer and takes
it to mean that the sun is the hub of a wheel, from where the beams spread
around like spokes.27 Modern authors tend to think that the wheels are like
bicycle tubes. One example is Krafft’s romantic drawing (see Figure 3.4).28
Another example is Brumbaugh, who describes them as “rings of hollow pipe
(a modern stovepipe gives the right idea).”29 Naddaf, in a recent article, also
draws such pipelike rings (see Figure 3.5).30 But are they right? Again, I think
that an anachronism is playing games with these authors. Let us take
Anaximander’s image seriously and look at some contemporary pictures of
(chariot) wheels (see Figure 3.6 a–c).31 They are much like coach wheels such
as those still used today. What we see are wheels, the cross-section of which
(i.e., of their rims) is a rectangle, not a circle or an oval, as in the tubelike
and concentrate on the globe. The only evidence is Diogenes Laertius, who
credits Anaximander with the construction of a sjaƒra. This word usually
is translated as “celestial globe.” Diogenes Laertius, being confused about the
shape of Anaximander’s earth, says that it is sjairoeid–V. Therefore, one
might wonder if by sjaƒra he did not simply mean a terrestrial globe.
Terrestrial globes (spherical maps of the earth), however, are of a much later
date. The earliest globes of the earth date from the time of Columbus’ dis-
covery of America,35 so Diogenes must have meant a celestial globe. But the
question arises whether he meant a globe on which the celestial constellations
were depicted, or a model in the sense of a so-called armillary sphere. A globe
depicting the celestial constellations would have been impossible, as it pre-
supposes knowledge of the stars of the Southern Hemisphere, which Diogenes
Laertius did not possess. So he must have meant an armillary sphere, which
is a model consisting of a number of brass rings representing the chief celes-
tial circles, such as the meridians, equator, ecliptic, horizon, and tropics. The
instrument probably was invented about 255 B.C.E. by Eratosthenes. The
question is, then, did Anaximander make a sphere, or any kind of three-
dimensional model of his universe? As we have seen above, it is not probable
that Anaximander taught the sphericity of the universe. Because of this, it is
likewise improbable that he made a spherical model (i.e., a sphere) of the
universe. Yet some authors fancy this kind of contraption, for example, Brumbaugh:
“Anaximander also built a model of the universe of stars and planets, with circling
wheels moving at different speeds. Like our modern planetarium projections, this
made it possible to speed up the observed patterns of planetary motion and find
in them a regularity and definite ratios of speed.”36 Of course, this is sheer fantasy.
I think it is safer to suppose that Diogenes Laertius again has been the victim
of the fallacy of anachronism, as he was not able to imagine a model of the
universe other than what he knew, namely, celestial globes (armillary spheres).
Therefore, we conclude that Anaximander did not make a sphere.
It is relatively easy to imagine what the plan perspective of
Anaximander’s universe would look like. More than a century ago, Diels
was the first to draw such a representation of the universe according to
Anaximander.37 But what would a three-dimensional model look like? I
think that the drawing or construction of such a model was beyond
Anaximander’s abilities. On the other hand, it is quite easy to explain the
movements of the celestial bodies with the help of a plan view, by making
broad gestures, describing circles in the air, and indicating direction, speed,
and inclination with one’s hands, as is said of a quarrel between Anaxagoras
and Oenopides.38 I will come back to the question of a three-dimensional
representation of Anaximander’s universe, where I will show that it is
possible for us to draw a three-dimensional picture of Anaximander’s
universe. He was such a great astronomer that such a picture is a true
rendition of the movements of the celestial bodies.
180 THE DISCOVERY OF SPACE
We may conclude that Anaximander was the best observer of the three.
According to the doxography, Anaximander placed the celestial bodies in
the wrong order: the sun is farthest from the earth, the stars (and planets)
nearest, and the moon in between.45 Scholars have wondered why Anaximander
could have been so ignorant to hold this strange order of the celestial bodies.
Let us first state that he was not the only one who made this mistake:
Metrodorus of Chios, a pupil of Democritus, and a certain Crates are said to
have held the same strange opinion, whereas Leucippus should have defended
the order of moon, stars, and sun.46 Apparently, something we consider per-
fectly obvious was not so obvious at all for these early astronomers. It was
Anaximander’s discovery that the celestial bodies lie behind one another, as
will be explained in due time. That he held the wrong order indicates that
this discovery was not as easy to make as we might think it was.
However, Anaximander’s order of the sun and moon is correct, so we
may ask how he obtained this knowledge. We would like to say that he could
have inferred it from an eclipse of the sun (e.g., the famous eclipse of May
28, 585 B.C.E., which Thales is wrongly said to have predicted). At that time,
Anaximander was about twenty-five years old. However, Anaximander’s theory
of the eclipses forbids such an inference. Eclipses, according to Anaximander,
are caused by the closing of the holes of the sun and the moon in their
wheels,47 and not by the relative positions of the sun and the moon to each
other and to the earth. Anaximander thought that the moon had a light of
its own,48 and that it shone through the hole in the moon wheel. Because of
this idea, the eclipses of the moon cannot be caused by the earth’s shadow,
thrown on it by the opposite sun. It must be the moon’s light itself, which
diminishes. In a similar way, when the moon is supposed to have a light of
its own, the regular phenomena of waxing and waning of the moon cannot
be explained by the way in which the light of the sun illuminates a bigger or
lesser part of the moon. At the time of a solar eclipse, on the other hand, it
is new moon or, in Anaximander’s words, the aperture of the moon wheel is
fully closed. This means that the darkening of the sun during the eclipse
cannot be ascribed to the moon disk being before the sun. Therefore,
Anaximander could not have concluded from the occurrence of eclipses that
the sun is farther away than the moon. Such a conclusion would presuppose
quite another and more modern theory of the true nature of the sun and
moon than was Anaximander’s. He must have had another criterion in mind
for deciding the order of the celestial bodies (viz., their respective brightness).
182 THE DISCOVERY OF SPACE
For some reason, he felt it natural to place the celestial bodies in the reverse
order of brightness. I come back to this issue later.
Some scholars wonder why Anaximander made the stars the nearest
celestial bodies, for he should have noticed the occurrence of star occultations
by the moon. I think that he is not too much to blame for the wrong position
of the stars. In contradistinction to Dicks,49 I think that the occultation of
stars by the moon is not so easy to observe (how many of us have ever
consciously witnessed the occultation of a star?) or, rather, that it could be
interpreted otherwise. Let us try to put ourselves in the position of someone
(Anaximander) who thinks—for whatever reason—that the brightest celestial
bodies are farthest away. Nowadays, we know that the stars are behind the
moon, and thus we speak of star occultation when we see a star disappear
behind the moon. But Anaximander had no reason at all, from his point of
view, to speak of a star occultation when he saw a star disappear when the
moon was at the same place. So it is a petitio principii to say that, for him,
occultations of stars were easy to observe. Perhaps he observed stars disap-
pearing and appearing again, but he did not observe—could not see it as—
the occultation of that star, for that interpretation did not fit his paradigm.
The easiest way to understand his way of looking at it—if he observed the
phenomenon at all—is that he must have thought that the brighter light of
the moon outshone the much smaller light of the star for awhile, and at the
time the moon was only a small, faint sickle, it was the sun whose light
outshone the star, for on those occasions, the moon was near the sun in the
sky. In addition, the fact that the stars looked much smaller than the sun and
the moon does not automatically lead to the conclusion that they are farther
away. In other words, those who, like Dicks, think that Anaximander is to
blame for not having understood the true meaning of star occultations are
victims of an anachronistic fallacy.
Let us now examine what an observer of the heavens sees with the naked
eye. When we look at night to the north (remember that Anaximander lived
on what we call the Northern Hemisphere), we see the stars circling around
a center that we call the North Pole of the heavens, which is approximately
the Polar Star. Near that center, the stars make full circles, that is, they never
set beyond the horizon. The farther away a star is from the North Pole, the
bigger its orbit, until we reach stars that set somewhere west of that pole and
rise somewhere east of it. The stars above our heads describe a part of a big
circle, before setting in the west. And when we turn to the south, we see stars
that make smaller and smaller arcs, and the most southern stars are above the
horizon for only a short time. We also see that the heaven of stars is slowly
changing from day to day, so that in the winter we see, in particular when
looking to the south, different stars than we do in the summer, and in the
spring we see other stars than those we see in the autumn.
Dirk Couprie 183
When we look at the heavens by day, we see the sun also describing part
of a circle from east to west. We may observe that sunrise is due east twice
a year, in the spring and autumn, whereas, at these very times of the year, the
sun sets due west.50 In the winter, the arc of the daily path of the sun is much
smaller, and the sun at noon stands lower and rises and sets further toward
the south. In the summer, on the other hand, its arc is much bigger, and the
sun at noon stands higher and rises and sets more toward the north. The
movement of the sun from east to west is somewhat slower than that of the
stars. This phenomenon accounts for the different starry skies in the different
seasons. We may observe that it takes one full year for the sun and stars to
attain the same positions with respect to each other again. The moon, at last,
we also see following a curved path from east to west, but it is remarkably
slower than the stars and the sun. It takes the moon approximately one
month to reach the same position among the stars again. During that month,
we see the arc of the daily path of the moon become bigger and smaller, more
northward and southward, just like that of the sun. And during that month,
we see the phases of the moon, from the full moon through the last quarter,
the new moon, first quarter, and the full moon again. Attentive observers can
even see that some stars, called the planets or wandering stars, make strange
movements among the others, but we will leave them out here, as the
doxography tells us nothing about Anaximander’s opinions of the planets. We
have to keep in mind this survey of what we can see with the naked eye when
we look at the heavens in order to understand the different features of
Anaximander’s astronomy.
FIGURE 3.7 The Author (left) and Hans Exterkate Measuring the Sun’s Shadow with a
Gnomon
taken as a kind of sundial.54 His photographs of the play of light and shadow
on the flutes of a single column and on a row of columns, in different seasons
and at different hours of the day, are fascinating. It reminds him of
Anaximander, who is said to have introduced the gnomon and to have erected
one in Sparta. He suggests that Anaximander and the architects discussed the
question of how to measure and arrange the columns to make them function
as gnomons. According to Kauffmann, the effect is particularly striking in the
case of columns that count twenty-four flutes. This is a weak point in his
hypothesis, as twenty-four fluted columns are rare. Moreover, as the play of
shadows on the flutes differs from season to season, this does not seem to be
an adequate way to tell the time. And, last but not least, as far as I know,
there exists no indication whatsoever in Greek literature that the Greeks in
daily life, or for whatever purpose, used the column as a sundial. When the
ordinary Greek in ancient times asked for the time, he or she did not look
at the columns of a temple. Instead, he or she could employ the oldest
gnomon by observing his or her own changing shadow on the ground. That
the Greeks were acquainted with this practice of indicating roughly the time
of the day is attested to in Aristophanes, where mealtimes are specified by the
length of the hungry person’s shadow.55
Dirk Couprie 185
The reports that say Anaximander was the first (Greek) to erect a gno-
mon, and that he constructed one in Sparta, are more trustworthy.56 This last
information gives us an indication of Anaximander’s reputation during his
lifetime. The practice of tilting the stick that casts the shadow, in order to
make it run parallel to the axis of the heaven (which equals, for that matter,
the earth’s axis) is a much later invention that, according to some,57 was made
as late as the first century A.D. This invention presupposes knowledge of the
sphericity of the earth, which Anaximander did not yet possess. The gnomon
has to be distinguished from the polos (p¬loV). The polos, which the Greeks,
according to Herodotus, also took over from the Babylonians,58 is an im-
proved version of the gnomon, in which the dial stick no longer stands on a
flat surface (the earth) but is erected in a hemisphere, the rim of which is
parallel to the horizon. This instrument presupposes the representation of the
universe as a sphere, which presumably was not Anaximander’s.
The gnomon is a simple instrument, but it is amazing how many differ-
ent observations can be made with its help. First, it is a device for telling the
time, although it always has to be remembered that “observations of the
shadow of a gnomon can give only the roughest indication of the time of day,
unless the gnomon is so placed that its axis is parallel to the axis of the
earth.”59 And this was, as said, not the case with Anaximander’s gnomon. The
main point of time to mark is noon. It can be done in different ways. Noon
is the time of day when the shadow of the gnomon is the longest, when the
sun is in the south. The shadow of the gnomon at noon lies on the meridian
of the observer, but this is not something Anaximander could have known,
as it presupposes knowledge of the sphericity of the earth. Another, more
precise, way of marking noon is by bisecting the angle between a morning
shadow and an afternoon shadow of the same length. This can be done by
marking the shadow of the tip of the gnomon at a certain time in the morn-
ing, and then drawing through that point a circular arc that is centered where
the gnomon touches the ground. In the afternoon one marks again where the
tip of the gnomon’s shadow touches the circular arc, and finally one bisects
the angle between these two shadow lines. This is shown in Figure 3.8, a plan
view, where G is the place where the gnomon stands, and GA and GB are
the two shadows in the morning and in the afternoon, so that GA = GB.60
The bisector (which, on a spherical earth, is the local meridian) will point
from the south to the north. An extra check is possible, as the lines AB and
NS have to make a right angle. This procedure already must have been
known in the Stone Age.61 The Egyptians already divided the arc of the
shadow, cast by the top of the gnomon from the earliest morning shadow till
the latest evening shadow, into twelve equal parts, denoting the hours of the
day. This resulted in “seasonal” hours, which are always one-twelfth of the
time from sunrise to sunset. Accordingly, the hours were of different lengths,
186 THE DISCOVERY OF SPACE
depending on the time of the year. The Egyptians obtained hours of constant
length by taking the equinoctial hours as a standard. In this system, which we
still use, the hours are of equal length, so that there are fewer hours in winter
from sunrise to sunset than there are in summer.62 Perhaps Anaximander was
acquainted with this practice.
The shadow of the gnomon at noon is shortest at the summer solstice
and longest at the winter solstice. The length of a year is the number of days
elapsing between two succeeding summer solstices (or winter solstices). The
angle from the tip of the gnomon to the ends of its shadows at the two
solstices can be measured and amounts to 47°, twice the inclination of the
ecliptic (see Figure 3.9).63 As we have seen, however, Anaximander probably
was not acquainted with the concept of the ecliptic, but he could have used
the angle as an indication of the height of the virtual cylinder, along which
his sun wheel would slide during the year, as will be explained later.
The dates of the equinoxes can be found in different ways. Sometimes
it is said that they can be found by noting at what day the earliest morning
shadow and the latest evening shadow make a straight angle (i.e., are dia-
metrically opposed).64 A better method is to observe the day when the path
Dirk Couprie 187
FIGURE 3.9 Measuring the Height of the Ecliptic with the Help of a Gnomon
of the shadow during the course of the day is a straight line, as is shown in
any book on sundials.65 A third method consists of bisecting the angle of the
shadows of the summer and winter solstices and noting the day when the
shadow of the gnomon hits that point on the ground (E in Figure 3.9). These
procedures show that Dicks is wrong when he writes that “the equinoxes
cannot be determined by simple observation alone.”66 None of these methods,
especially the second, presupposes knowledge of the obliquity of the ecliptic.
The word “equinox” (Îshmer√a) simply means that two times a year night
and day are equal in time. The concept of the equinoxes is not “a more
sophisticated one, involving necessarily the complete picture of the spherical
earth and the celestial sphere with equator and tropics and the ecliptic as a
great circle,” as Dicks wrongly supposes, nor are these ideas “entirely anach-
ronistic for the sixth century B.C.”67 The curves that the shadow of the top of
the gnomon casts on all other days except the equinoxes are hyperbolic (at
Anaximander’s latitude, as at all latitudes below 66°),68 with both solstitial
curves as the extremes on both sides of the equinoctial line, as shown in
Figure 3.10.69 Anaximander could have drawn these lines without knowing of
course that these curves were hyperbolas. The observations described above
188 THE DISCOVERY OF SPACE
FIGURE 3.10 Equinoctial Line and Solstitial Curves on a Roman Sundial (1st or 2d Century B.C.E.)
FIGURE 3.11 Measuring the Observer’s Latitude with the Help of a Gnomon
Measured by a gnomon, the azimuth of the sun is the angle between the
shadow, thrown by the sun at a certain moment and the line of its shadow
at noon (see Figure 3.12).71 In this way, Anaximander easily could have mea-
sured the azimuth of sunrise and sunset at the days of the solstices. We will
come back to this when we discuss Anaximander’s map of the world.
So far we have treated methods of using the gnomon for making obser-
vations on the position and movements of the sun across the celestial dome. As
they have nothing to do with the notion of depth in the universe and the
distance of the celestial bodies from the earth, these applications were of no use
for Anaximander’s specific achievement in astronomy, that is, for what I call his
discovery of space. However, there are ways of using the gnomon that he could
have used in the context of his theories about the free-floating earth and the
distances of the celestial bodies, of which, apparently, he did not know.
In order to understand the measurements that Anaximander could have
made, we have to start a few centuries later, when Eratosthenes (276–194
B.C.E.), maintaining that the earth was spherical, made use of a gnomon in
order to measure the circumference of the earth. He had read that at the date
190 THE DISCOVERY OF SPACE
of the summer solstice in Syene (now Aswan), the sun at noon stood exactly
in the zenith, so that a gnomon did not cast any shadow. He knew, however,
from observations with a gnomon, that at the same date and time in Alex-
andria the sun was 7° off the zenith. According to Eratosthenes, the sun is
so far away that its rays run parallel to one another, even at such a distance
as that between Syene and Alexandria. So the difference between the two
Dirk Couprie 191
shadow angles had to be caused by the curvature of the earth, and the earth’s
circumference must be 360/7 = about fifty times the distance between Syene
and Alexandria, which was 5,000 stadia, and thus 250,000 stadia (see Figure
3.13).72 Assuming that a stadium equals 157.5 meters, this results in a cir-
FIGURE 3.13 Eratosthenes’ Method of Measuring the Circumference of the Earth with the
Help of Two Gnomons
192 THE DISCOVERY OF SPACE
FIGURE 3.14 How to Measure the Distance of the Sun with the Help of Two Gnomons
When You Think that the Earth is Flat
Dirk Couprie 193
gnomon AB to its shadow AX is 10:2 = 5:1. The distance AC = 1,000 li. Since
for every thousand li southward, the shadow diminishes by 1 inch, the Chinese
astronomers concluded that at a distance of 20,000 li, the gnomon does not
throw a shadow at all: AT = 20,000 li. Now the proportions of the triangle
XAB are the same as those of the triangle XTS. So is TS:TX = AB:AX = 5:1,
thus TS must be 5 20,000 = 100,000 li, which is about 50,000 kilometers.
Of course, this result is completely wrong (the real mean distance of the
sun being 150,000,000 kilometers), because it starts from the wrong hypoth-
esis of a flat earth, so it does not account for its curvature from one obser-
vation post to the other. But this does not alter the fact that Anaximander
could have used this method to estimate the distance of the sun. The Chinese
method is essentially the same as Thales must have used to estimate the
distance of a ship at sea,75 illustrated in Figure 3.15,76 where an observer
stands on the top of a tower of known height, holding a horizontal rod, to
see the ship in direct line with a point on that rod. Now the triangles EBS
and EHP are similar, and EH:EB = HP:BS; as EH, EB, and HP are known,
BS can be easily calculated.
As we know, according to Anaximander, the distance of the sun is 27
earth diameters. There is, however, no indication whatsoever that he derived
this from any kind of measuring. On the contrary, in Anaximander’s case, the
Chinese method amounts to an earth diameter of 50,000:27 = ± 1,850 kilo-
meters (N.B., this concerns the diameter of a flat, not of a spherical earth).
FIGURE 3.15 How Thales Could have Measured the Distance of a Ship at Sea
194 THE DISCOVERY OF SPACE
This result is too small for Anaximander, as the greatest known distance at
his time, between Babylon and the Pillars of Hercules, is already about 5,000
kilometers. Therefore, we may conclude that if Anaximander had made use
the Chinese method, he never would have said that the distance of the sun
equals 27 earth diameters. Moreover, if Anaximander really had measured the
distance to the sun, the doxographers certainly would have reported it.
defining this shape and these limits. Therefore, it must be the inhabited part
of the earth (oÎkoum°nh) that is meant. The procedure would be like this:
Put a circular model (e.g., a column drum) on a horizontal floor. Erect a
gnomon at the center of your model and mark where the shadow line from
the point of sunrise at one of the equinoxes cuts its circumference. Do the
same for the equinoctial sunset. Connect the two points you have found. This
line divides the model into two equal halves. Repeat this procedure for the
northernmost points of sunrise and sunset at the summer solstice, and for the
southernmost points of sunrise and sunset at the winter solstice. Heidel called
the three parallel lines thus found the “Ionian equator and tropics.”84 From
the point of view of ancient astronomy, this is the most interesting part of his
book. “Equator” and “tropics” are, of course, concepts that belong to a spheri-
cal earth and universe, but Heidel made it feasible for them to have a kind
of meaning on a flat earth as well. The locations of these “equator and tropics”
are completely different from those of the equator and tropics on a spherical
earth. Heidel concludes that “we have, then, three clearly indicated parallels
on the Ionian maps, corresponding to the tropics and the equator on our
maps, but drawn at places where we should not think of locating them.”85
Now we can start to fill in our model. First we draw the surrounding
Ocean by circling its inner border with a compass. We mark the center of our
model and call it “Delphi,” the world’s navel. We have to realize that on a flat
earth, the climate must be thought of as being colder the farther one goes to
the north, and warmer the farther one goes to the south. Therefore, the
“equator” through Delphi divides the earth into two equal halves, a northern,
colder half, and a southern, warmer half. The next site we mark on the
equator is the strait between the Pillars of Hercules, which connects the
Mediterranean Sea to the encircling Ocean. We may suppose that the daring
Milesian sailors had come so far. Heidel says that Polybius, making use of
ancient maps, placed the Pillars of Hercules at the “equinoctial sunset.”86 The
line from Delphi to the Pillars of Hercules is the radius of the round earth
within the ring of the Ocean. At an appropriate distance, east of Delphi, we
make the “equator” run through Miletus, as Brumbaugh, Lee, and Heidel do
on their sketch maps.87 In reality, of course, the Pillars of Hercules, Delphi,
and Miletus do not lie exactly on the same east-west-line (we would say on
the same line of latitude), but the differences are so small that we are entitled
to suppose that the early mapmakers did not know this. Now we are able to
draw the contours of the Mediterranean (and the Euxine). We draw both the
Caspian Sea and the Red Sea as bays of the Ocean.88 The area between the
two “tropics” is the oÎkoum°nh with its oblong shape that forms, according
to Heidel, the frame of the ancient Greek maps. It includes the countries
around the Mediterranean: the Iberian, Italian, and Balkan peninsulae, a
small region to the north of these countries, the Euxine, Asia Minor,
196 THE DISCOVERY OF SPACE
B.C.E. (Figure 3.18).95 The goddess of the heaven, Nut, arches over the world,
which is supported by two arms, making the “ka”-sign. We observe again an
encircling Ocean. Inside it is a ring representing the names of the foreign
countries; then follows a ring with hieroglyphs of the Egyptian nomes (dis-
tricts); finally, in the innermost circle, the underworld is represented. An-
Dirk Couprie 199
other, unfortunately rather mutilated and difficult to date (but probably also
late), example of such a map is described by Clère (Figure 3.19).96
These maps can be characterized as a mixture of mythological and sym-
bolic or schematic features. The triangular “mountains” outside of the Bitter
River on the Babylonian map and the goddess, Nut, arching over the earth
200 THE DISCOVERY OF SPACE
on the Egyptian map are essential features of these maps, but they clearly
have no geographical meaning. Moreover, these maps do not show borders or
seas but only a few schematic lines (especially on the Babylonian map) and
a list of geographical names (especially on the Egyptian map). The difference
between Anaximander’s map, on the one hand, and the Babylonian and
Egyptian maps, on the other hand, is analogous to the difference between
Dirk Couprie 201
only arcs in reality also describe full circles, just like those near the Polar Star.
Regarding the sun and the moon, we can observe that the arcs they describe
are sometimes bigger and sometimes smaller, and we are able to predict
exactly where they will rise the next day. Therefore, it seems not too bold of
a conjecture to say that these celestial bodies also describe full circles. Nev-
ertheless, it was a daring conclusion, precisely because it necessarily entailed
the concept of the earth hanging free and unsupported in space.
In Homer and Hesiod, the sun, moon, and stars rise from the ocean in
the east and plunge into it again in the west. But, as Dreyer remarks, “what
becomes of the heavenly bodies between their setting and rising is not stated,
but since Tartarus is never illuminated by the sun, they cannot have been
supposed to pass under the earth.”98 In other cultures, we occasionally meet
a representation of, for example, the sun being transported by a boat through
the waters underneath the earth. These ideas are quite different from
Anaximander’s. How daring his conception was is shown by the fact that his
follower, Anaximenes, taught that the celestial bodies do not pass beneath the
earth but go around it “like a felt hat turns about our head.”99 Anaximander
compared the full circles of the celestial bodies with chariot wheels. They are
made of thick air and filled with fire. At one place, such a celestial wheel has
a hole in it, through which the inner fire shines. At first sight, this looks like
a fantastic image, sprung from a bizarre mind. I have already said, however,
that I will take this image seriously, and in the next paragraph, I argue that
there are good reasons for it.
Anaximander boldly purported the idea that the earth floats free in the center
of the universe. This idea marks a complete revolution in man’s understanding
of the universe. How revolutionary this conception was is again illustrated by
Anaximenes, who apparently found the idea too daring and let the earth float
like a lid on the air. Obviously the earth hanging free in space is not some-
thing Anaximander could have observed. More than 2,500 years later astro-
nauts really saw the unsupported earth floating in space and thus provided
the ultimate confirmation of Anaximander’s conception (see Figure 3.20).100
We may assume that Anaximander somehow had to defend his bold
theory of the free-floating, unsupported earth against the obvious question of
why the earth does not fall. The tradition gives two versions of Anaximander’s
argument, in which he gives the reason the earth does not fall, the oldest of
which is in Aristotle: “But there are some who say that it stays where it is
because of equality, such as among the ancients Anaximander. For that which
is situated in the center and at equal distances from the extremes has no
Dirk Couprie 203
FIGURE 3.20 The Ultimate Proof of Anaximander’s Conception of the Earth Hanging Free
in Space
the earth, both of which do not belong to Anaximander’s teachings. The first
reason is not very strong. The same argument as Anaximander’s is ascribed
by Aetius to Parmenides and Democritus, the first of which was born only
twelve years after Anaximander’s death.104 Anyway, historical isolation as such
is no proof. In the history of Greek astronomy, another famous example of
historical isolation exists in the third century B.C.E. in Aristarchus’ teaching
of the earth going around the sun. The second objection, however, is more
serious, so we will look at it somewhat more closely. In order to do this, we
also will have to take into consideration the famous text in Plato’s Phaedo,
where Anaximander’s argument returns: “Now, however, I am persuaded, he
said, that, in the first place, if the earth, being round, is in the middle of the
heaven, it needs neither air nor any other suchlike constraint in order not to
fall, but the all-sided equality of the heaven to itself in all respects and the
equilibrium of the earth itself are sufficient to hold it still; for a thing in
equilibrium set in the center of something identical will have no reason to
incline to one side rather than another, but being neutral it will remain
immobile.”105 I agree with those who assume that Plato here, for the first
time in history, tries to express the sphericity of the earth (strogg§lh and
perijer–V, meaning sjairoeid–V), using the image of a many-colored ball
of twelve pieces of leather,106 and taking the sphericity of the universe for
granted.107
As Anaximander was not acquainted with the sphericity of the uni-
verse108 and of the earth, it is not so easy to understand what the expression
“at equal distances from the extremes” means. Perhaps one part of the solu-
tion is that his argument somehow has been confused with an instruction for
making a map of his universe, as will be explained in the next section. It is
tempting to see Anaximander’s map of the universe109 as a visual and math-
ematical confirmation of the argument. On such a map, the earth appears
naturally in the center, at equal distances from the rings of the sun, moon,
and stars. However, being a two-dimensional map, it can at best illustrate
why the earth does not drift off sideways. To the most important problem,
namely, why the earth, being unsupported, does not fall, this illustration
provides no answer at all. Also the suggestion that Anaximander must have
thought of the wheels of the stars making a globe together will not do. In the
argument, there is talk of “the extremes,” whereas, according to Anaximander,
the stars are closest to the earth.
Aristotle, in his reproduction of Anaximander’s argument, apparently
leans heavily on Plato’s text and has a tendency to completeness, of which one
may doubt whether it reflects anything that Anaximander could have said.
The argument, according to Aristotle, not only answers the question why the
earth does not fall but also why it does not rise up, and why it does not drift
off sideways. The last sentence in Aristotle’s version is another expression of
Dirk Couprie 205
same kind of argument a hair, which was subject to an even pulling power
from opposing sides, would not break, and that a man, being just as hungry
as thirsty, placed in between food and drink, must necessarily remain where
he is and starve. The latter argument is virtually the same as that which since
the Middle Ages has been known as “Buridan’s ass.” The great protagonist of
the principle of sufficient reason, Leibniz, however, uses an example, which
he ascribes to Archimedes, and which reminds us strongly of the way Aristotle’s
Anaximander uses it: “And therefore Archimedes (. . .), in his book De
aequilibrio, was obliged to make use of a particular case of the great Principle
of a sufficient reason. He takes it for granted that if there be a balance in
which everything is alike on both sides, and if equal weights are hung on the
two ends of that balance, the whole will stay at rest. This is because there is
no reason why one side should weigh down, rather than the other.”115 This
very argument is not extant at all in Archimedes’ book,116 but it shows the
same structure as those of Anaximander and Plato, to wit a negative use of
the principle accounting for the existence of a situation of equilibrium. In a
similar way, Leibniz uses the Principle in his famous arguments against ab-
solute space and time: “Now from hence it follows (supposing space to be
something in itself, besides the order of bodies among themselves) that it is
impossible there should be a reason, why God, preserving the same situations of
bodies among themselves, would have placed them in space after one particu-
lar manner and not otherwise, and why everything was not placed the quite
contrary way (for instance) by changing East into West.”117 And with time:
“Suppose that someone should ask why God did not create everything a year
sooner, and that the same person should infer from thence that God has done
something concerning which it is not possible there should be a reason why he did
so and not otherwise. One should answer him that this inference would be
right if time was something distinct from things existing in time, for it would
be impossible that there were any reasons why things should be applied to
such particular moments rather than to others, their succession remaining the
same.”118
One may question, however, whether these arguments may count as
applications of the Principle itself. The arguments, from Anaximander to
Leibniz, argue from the nonexistence of a sufficient reason. This is a dangerous
way of arguing, for it contains a kind of subjective aspect: “as far as I know.”
Absolute propositions concerning the nonexistence of things are always in
danger of becoming falsified upon closer investigation. Anaximander’s case is
a good example of the consequences of this danger, because closer investiga-
tion has taught us that the earth is not the center of the universe, so the
argument as such collapses. The Principle of Sufficient Reason itself, how-
ever, says something positive, namely, that everything has a reason that it is
so, and not otherwise. Unfortunately this is obscured by Leibniz’ tendency to
Dirk Couprie 207
The very theory of the celestial bodies making full circles, which entails
that the earth must be floating free in space, also brings forth the solution to
the problem of why the earth does not fall. If the celestial bodies had been
free-floating themselves—as we now know that they are—Anaximander’s
argument would not have made sense. This must have been one of the rea-
sons he conceived of the celestial bodies as wheels: each spot on a celestial
wheel has an equal distance to the earth as any other. Let us look at the same
problem from the other side as well. There is no doxographic evidence of it,
but it is quite certain that the question, why the celestial bodies do not fall
upon the earth, must have been as serious a problem to Anaximander as the
opposite question, why the earth does not fall. The explanation of the celestial
bodies as wheels, then, provides an answer to both questions: just like the
earth, the celestial bodies have no reason whatsoever to move otherwise than
in circles around the earth, as each point on them is always as far from the
earth as any other.125 It is because of reasons such as this that, for ages to
come, when Anaximander’s concept of the universe had been replaced by a
spherical one, the celestial bodies were thought of as somehow being attached
to sphere shells, and not as free-floating bodies. In fact, we have to wait until
Newton’s law of gravity for a more convincing answer to the question of why
the celestial bodies do not fall. These considerations give the tradition on
Anaximander’s wheels more trustworthiness than Dicks credits it with.126
The ideas of the free-floating earth and the celestial bodies as wheels making full
circles around the earth are the first marks of Anaximander’s new concept of the
universe, but they need the addition of the third, namely, the theory of the
different distances of the celestial bodies to the earth. When Anaximander looked
at the heavens, he imagined, for the first time in history, space. His vision implied
depth in the universe, that is, the idea that the celestial bodies lie behind one
another. Although it sounds simple, this is a remarkable idea, because it cannot
be based on direct observation. We do not see depth in the universe. The more
natural and primitive idea is that of the celestial vault, a kind of dome or tent,
onto which the celestial bodies are attached, all of them at the same distance, like
those in a planetarium. One meets this kind of conception in Homer, when he
speaks of the brazen or iron heaven, which apparently is conceived of as some-
thing solid, supported by Atlas and pillars.127 The cosmology of Hesiod, Pindar,
and Theognis is founded on similar ideas.128 Anaximander, on the contrary, “had
shaken himself free of the old idea that the heavens are a solid vault.”129 How
revolutionary this concept was is shown by Anaximenes, who reintroduced the
image of the celestial vault by teaching that the stars are like nails driven into, or
like fiery leaves painted onto, the crystalline celestial vault.130
Dirk Couprie 209
As far as I can see, it is only the lowermost heaven in this text, which
is associated with heavenly bodies painted on it, not unlike Anaximenes’
ideas. The other two layers apparently have a more religious meaning. This
becomes clearer when we see that in the next lines of the text, the three layers
of the heaven are mirrored by three layers of the earth: the upper earth, where
mankind, frail like a gust of wind, lives; the middle earth, where He (Marduk)
settled his father Ea; and the lower earth, where the 600 Anunnaki dwell. We
210 THE DISCOVERY OF SPACE
may conclude that in the neighboring civilizations there exists hardly any
evidence, before and during Anaximander’s time, of a conception of the heav-
ens that can be compared with his.
Regarding Anaximander’s conception of the universe, I fully agree with
the words of Heidel: “(. . .) the solid earth, still regarded as the disk, sug-
gested by the horizon, was thought to stand still, while the outer bands,
composed of mist and fire, continue to revolve about it. This was a very bold
hypothesis. While it saved appearances in respect to the earth, it did away, at
a stroke, with the notion of the ‘inverted bowl’ or hemisphere of the sky.”136
Perhaps it is not too daring to read into one text in his cosmogony a remi-
niscence that Anaximander was fully aware of the fact that he was exploding
the old conception of the universe. It is reported that originally a sphere of
fire, originating from the g¬nimon, grew around the earth like the bark of a
tree, before falling apart into separate rings for the sun, moon and stars.137
When we take the tree as the customary metaphor for the heavenly tree,
which is the axis of the heavens, the bark of that tree can be regarded as an
image of the celestial vault that explodes as a kind of heavenly fireworks,138
the result of which is a system of concentric rings.
In order to get a better understanding of Anaximander’s idea of depth in
the universe, let us look again at his strange order of the celestial bodies. We
tend to say that, although we cannot see that the celestial bodies are behind
each other, we may infer that the sun is farther away than the moon by
observing an eclipse of the sun, and we may infer that the stars are far away,
because they are much smaller than the sun and the moon, or because we can
observe what we call star occultation. Anaximander, however, as we saw ear-
lier in our discussion of anachronistic fallacy, did not make either of these
inferences. Our conclusion must be that neither could Anaximander have
observed the spatial relations of the celestial bodies, nor could he have inferred
these relations in the way we are used to. So there must be another reason for
his putting the celestial bodies in this strange order.
The inner logic of Anaximander’s order of the celestial bodies seems to
be their brightness: the farther away, the brighter they are.139 Krafft points to
the curious fact that in German it is idiomatic to say “Sonne, Mond und
Sterne” when enumerating the celestial bodies.140 The same idiomatic expres-
sion also occurs in English: “sun, moon, and stars,” in this order, which is
Anaximander’s order and not the real one. In daily language, we still use the
order of decreasing brightness. It is like an Anaximandrian fossil in our
language. Some scholars have made the suggestion that the original sphere of
fire that originated from the g¬nimon, grew around the earth like the bark of
a tree, and fell apart into the separate rings of the sun, moon, and stars still
resides at the periphery of Anaximander’s universe.141 This fire, they suppose,
remains invisible, hidden by the air that also hides the fire within the celestial
wheels, except where the openings are, through which we see the fire as the
Dirk Couprie 211
sun, moon, or stars. The farther away from the surrounding fire, the less
bright (and hot) the celestial bodies are, and this should explain the order of
the celestial bodies. Sometimes the suggestion has been made that Anaximander
borrowed his order of the celestial bodies from shamanistic rituals of the
voyage of the soul through the spheres of the stars, moon, and sun to the
realm of the heavenly light.142 The outer heaven, then, should have been
identified with this heavenly realm. West even suggests the number 36 as the
diameter of this “outer o¶ran¬V.”143 The idea of a surrounding fire outside of
the sun wheel, however, must be judged as speculative, as there exists no
textual evidence for it, unless one identifies it with Anaximander’s Òpeiron
which, according to Hippolytus, surrounds all “worlds” (k¬smoi).144
Bodnár has tried to offer a kind of “physical” explanation for the order
of the celestial bodies in which there is no need for such a surrounding fire:
the brighter lights can more easily penetrate the misty envelopes of the inside
rings.145 Indeed, this solution tries to explain how it is possible that we see the
light of the outer celestial bodies, but it does not explain why the order of the
celestial bodies is as it is, according to Anaximander. Perhaps an inference
might be drawn from a text in the doxography on Leucippus, who says that
all of the celestial bodies are inflamed by the velocity of their movement.146
One might suppose that the faster a celestial body moves (i.e., the bigger its
wheel is), the hotter and brighter it is. Using the numbers found in the next
section, taking the diameter of the earth as 5,000 km, and p = 3, the velocity
of the sun wheel amounts to: 2 3 (27 5,000): 24 = 33,750 km/hour,
which must have been quite an amazing speed in the eyes of the ancient
Greeks.147 Considerations like those dealt with above could have led
Anaximander to his order of the celestial bodies, but none of the explanations
given is completely convincing. Nevertheless, the conclusion seems to be that
somehow it must have been the differences in brightness between the celestial
bodies, which made Anaximander think that they lay behind one another.
ANAXIMANDER’S NUMBERS
AND A MAP OF HIS UNIVERSE
among the stars, at the same distance from the earth.151 However, a number
for the stars has not been handed down. Building on the work of earlier
scholars such as Tannery and Diels, O’Brien describes the distances of the
celestial bodies in Anaximander’s universe as being based on calculations that
were simple numerical proportions.152 According to this reconstruction, the
diameters of the inner and outer rim of the sun wheel measure 27 and 28, the
inner and outer diameters of the moon wheel 18 and 19, and those of the
stars, which are nearest to the earth, 9 and 10 earth diameters. Others have
argued that the numbers must refer not to the diameters but to the radii of
the rings, that is, their distances to the earth. I think that they are right, for
three reasons. The first reason has to do with a problem that has been delin-
eated by Kirk.153 It is recorded that, according to Anaximander, the sun is the
same size as the earth.154 This means, Kirk assumes, that the width of the sun
ring must be the same as the diameter of the earth. But this would result in
a difference of two earth diameters between the diameter of the inner rim of
the sun wheel and the diameter of its outer rim instead of one. The same
holds mutatis mutandis for the other rings. O’Brien tries to solve this problem
by maintaining that “there is no reason why we should not think of the
thickness or width of the rim of the sun wheel as equal to one half of the
earth’s diameter.”155 Although he is right when he says that “this will give a
difference of one between the inner and the outer diameter of the sun wheel,”156
he does not explain how to press a whole sun into a ring that measures half
of its width. Thus he leaves Kirk’s problem unanswered.
The second reason I think the numbers must indicate radii has to do
with what I call Anaximander’s main discovery, namely, that the celestial
bodies are at different distances from the earth. In contradistinction to diam-
eters, radii indicate distances (from the celestial wheels to the central earth),
and these were exactly what Anaximander was interested in. The third reason
is that I tend to read the numbers as a kind of instruction for drawing a map
of the universe. Anaximander’s conception of the universe presupposes the
possibility of elevating oneself in thought from the earth to a virtual point of
view from which one could survey the universe as a whole. This is the point
of view of a mapmaker. As we have seen, Anaximander was the first to make
a mappa mundi, a map of the world. I take it that where the doxography
anachronistically speaks of a “sphere,”157 these reports go back to a map of the
universe, drawn by Anaximander. Perhaps even the words ˙lwV gewmetr√aV
•pot§pwsin ⁄deixen158 must be understood as a reminiscence of such a map.
If we, then, suppose that Anaximander made a map of his universe, he must
have used measures in order to make his drawing. In other words, somehow
Anaximander’s numbers also must have played a role in his attempt to visu-
alize his universe. The instruction, then, reads somewhat like this: “Take a
compass and draw a little circle; this is the earth. Call its diameter 1 unit.
Dirk Couprie 213
Now leave one of the legs of the compass in the center, put the other leg at
a distance of nine units and draw a circle; this is the inner rim of the star ring.
Now put the same leg of the compass at 10 units; this is the outer rim of the
star circle, etc., etc.” Moreover, I do not only agree with Kirk that the clause
'Anax√mandroV t¿n mÆn ≠\lion ≥son eΔnai t˛› g˛› means that the width of
the sun ring must be the same as the diameter of the earth, but I tend to see
it also as a kind of drawing instruction: “make sure that the width of the
ring(s) is one earth-diameter.” As said earlier, the argument that the earth
does not fall “because of its equal distance to the extremes” might be under-
stood as well as a distorted drawing instruction. My version of a map of
Anaximander’s universe is shown in Figure 3.21.159
instruments and geometry.166 This is the main reason I think that O’Brien’s
terminology is misleading when he speaks of “Anaximander’s measurements,”
and why I prefer the word “numbers.” That the celestial bodies are behind
each other is not something we can see, let alone that these distances could
have been measured in Anaximander’s time.
Anaximander had a preoccupation with wheels, as shown in his description
of the celestial bodies as wheels. The doxography tells us that the earth, according
to Anaximander, resembles a column drum.167 He could have derived the ratio
1:3 between the height of the earth and its diameter from that kind of building
brick, sometimes showing roughly those dimensions (see Figure 3.22).168 But to
say that the proportions of the wheels, being simple multiples of 3 (+ 1 for the
width of each wheel), should reveal a mathematical genius is both exaggerated
and question begging, for it does not explain how he got these numbers.
Perhaps we may understand a bit more of Anaximander’s universe when
we follow a suggestion made by West and Krafft.169 They quote Hesiod’s
Theogony (722–725), where it is said that a brazen anvil (according to Krafft,
a meteorite) would take nine days to fall from Heaven to Earth, and again
nine days to fall from Earth to Tartarus. Or, more precisely, it takes nine days
to fall, arriving on the tenth day.170 It is not a bold guess to suppose that
Anaximander must have known this text. The agreement with his numbers
is too close to neglect, for the numbers 9 and 10 are exactly those extrapolated
by Tannery for Anaximander’s star wheel. Moreover, according to Krafft, in
the Greek counting system Hesiod’s numbers should be taken to mean “a very
long time.” Thus Troy was conquered in the tenth year after having stood the
siege for nine years; and Odysseus scoured the seas for nine years before
reaching his homeland in the tenth year.171 Hesiod can be seen as a forerun-
ner to Anaximander, for he tried to imagine the distance to the heaven, and
even to the depths underneath the earth, although he apparently did not
link these distances to the celestial bodies, let alone to celestial bodies at
different distances. We may infer that Anaximander, with his numbers 9
and 10 for the star ring, simply was trying to say that the stars were very
far away. Then the other numbers can easily be interpreted as “farther” (for
the moon ring) and “farthest” (for the sun ring). And this is exactly what
we should expect one to say who has discovered that the image of the
celestial vault is wrong, but that the celestial bodies are behind one another
and who wishes to share this new knowledge with his fellow citizens. This
solution explains why Anaximander used the numbers that have been handed
down to us: he wished to express, by using numbers that his readers could
understand as such, that the stars, the moon, and the sun are far, farther,
and farthest away.
Accepting Anaximander’s numbers, the difficulty is, however, how to
arrive at a satisfying map of the universe that is not in flat contradiction with
216 THE DISCOVERY OF SPACE
the most obvious observational phenomena. The assumption that the num-
bers are not based on observation does not mean that the numbers have
nothing to do with observational data at all. Any interpretation entailing
unacceptable observational consequences that were easy for Anaximander
himself to observe must be wrong. In other words, Anaximander’s numbers
cannot be in flagrant discrepancy with such observational data, for otherwise
he would have noticed it. Here again we must try to look at the celestial
phenomena as Anaximander must have done.
As it is recorded that, according to Anaximander, the sun is the same size
as the earth, I take this to mean that the width of the wheel of the sun (or
its ring on a map) is 1 earth diameter, and that the same holds true for the
other celestial wheels. This would mean that if the numbers were to refer to
distances, then the distance of the inner rim of the sun wheel measures 28
and the outer rim of the sun wheel measures 29 earth diameters.172 However,
according to this interpretation, we would get a sun wheel of 162 (2p (= 3)
27) suns which, put one after another, make up the total sun ring,173 and
we would get, accordingly, an angular diameter of some 2°15' for the sun,
which is about 4.5 times too big.174 In reality, some 720 suns one after an-
other, make up a full circle, as the sun’s actual angular diameter is approxi-
mately 30'. The outcome of this calculation is so far out of agreement with
the most obvious observational data that it must have struck Anaximander’s
eye.175 Even he would have been able to see that his sun ring (the daily path
of the sun) was much bigger than 162 suns, put one after another. This is
even more true if we can trust the report that Thales had already discovered
that the apparent diameter of the sun is 1/720 times its orbit.176 In other
words, even when the numbers refer to the radii of the celestial wheels, the
radii indicating distances, the problem of the angular diameter of the sun
remains. So we are left in need of a solution that would make it possible to
depict Anaximander’s universe without contradicting the most obvious obser-
vational data, in particular, the angular diameter of the sun.
Before trying to offer a solution, we must take into account another com-
plication, namely, the moon. In the doxography, nothing is said about its mag-
nitude, but I think we may assume that Anaximander could have said that the
moon was the same size as the earth too. If we follow Kirk, this would mean
that the width of the moon ring equals the diameter of the earth as well. In
fact, all of the maps that scholars have tried to draw seem to presuppose it.
This, however, seems to mean that it would need 108 moons to complete the
full circle of the moon ring, and thus an angular diameter for the moon of more
than 3°15'. These results are an even more flagrant contradiction of the visual
phenomena than those relating to the sun, for the moon, as it appears in the
sky, is approximately as big as the sun, so its angular diameter also is about 30'.
218 THE DISCOVERY OF SPACE
of the ecliptic, and who all make the same characteristic mistake. Uninten-
tionally, they show how difficult it is to place oneself in Anaximander’s po-
sition and look at the skies as though the earth were flat. The case is best
illustrated by a drawing made by Krafft (see Figure 3.24).180 This picture is
one of the two options presented by Krafft as an explanation of the move-
ment of the sun wheel in Anaximander’s universe, which he takes to be the
220 THE DISCOVERY OF SPACE
FIGURE 3.24 Wrong Picture of the Movement (“Oscillation”) of the Sun Wheel
right one, as is seen from his criticism of the other and from his rather
romantic representation of Anaximander’s universe, which also is wrong (refer
back to Figure 3.4). What Krafft has tried to draw has been expressed by
Sarton: “The inclination of that plane [viz., the ecliptic, D.C.] to the horizon
varied from day to day, being smallest at the winter solstice (. . .) and largest
at the summer solstice (. . .); the plane reached its halfway position at the
times of the equinoxes.”181 Both Krafft and Sarton took this mistaken idea
from Heath who, in his turn, borrowed it from Neuhäuser. I have criticized
their drawings and explanations elsewhere.182 Krafft draws the movement of
the ecliptic as a kind of oscillation (the word already used by Neuhäuser)
between two fixed points at the horizon, obviously meaning east and west.
This would mean that the sun is always at the same place at the horizon,
which is not the case, for in summer the sun rises and sets farther to the
north and in winter farther to the south.183 In reality, the movement of the
ecliptic, as seen from the earth, is quite different. This is best described, in
the words of Elisabeth Mulder, as a dance: the ecliptic dances around the
earth when the celestial bodies accomplish their daily rotations. Her pictures
of it are shown in Figure 3.25.184
The ecliptic is the annual circular path of the sun from west to east across
the stars. There is, however, another circular path of the sun that is quite easy
Dirk Couprie 221
to discern, namely, its daily path from east to west. The most plausible sup-
position is that this orbit coincides with Anaximander’s sun wheel. The same
holds mutatis mutandis for the moon. In the doxography, almost nothing
about Anaximander’s opinions about the stars has been handed down to us.
But probably the best way to imagine them is as a conglomerate of several
wheels, each of which has one or more holes, through which the inner fire
shines, which we see as stars. The most likely sum total of these star wheels
is a sphere, but it also is possible to imagine them as making up a huge
cylinder together (or perhaps even a cylinder of infinite length).185 In either
case, the only movement of these star wheels is a rotation around the earth
from east to west, always at the same speed, and always at the same place
relative to one another in the heavens. The meaning of the last suggestion—
the star wheels forming a huge cylinder together—may become more plau-
sible when we look at a modern photograph, a time exposure of the stars,
circling around the northern celestial pole: it is as if we look into a large
cylinder (see Figure 3.26).186 Perhaps one would object that to elucidate
Anaximander’s astronomy with the help of a photograph is a flagrant anach-
ronism. In this case, the objection is unfounded, because the camera knows
nothing about astronomical theories concerning which celestial body turns
around which, but it only registers what it sees.
222 THE DISCOVERY OF SPACE
FIGURE 3.26 The Star Wheels Making up a Cylinder, Visualized by a Modern Photograph
(Compare with Figure 3.28b)
Dirk Couprie 223
The sun wheel shows the same rotation from east to west as the stars,
but there are two differences. The first is that the speed of the rotation of the
sun wheel is not the same as that of the stars. We can see this phenomenon
by observing how the sun lags behind by approximately 1° per day. Today we
describe this movement of the sun (and mutatis mutandis of the moon and the
planets) as a retrograde movement, from west to east, which is a counter-
movement to the daily rotation from east to west. In terms of Anaximander’s
ancient astronomy, it is more appropriate and less anachronistic to describe it
as a slower movement of the sun wheel from east to west. The result is that
we see different stars in different seasons, until the sun, at the end of a year,
assumes its old position between the stars. The second difference is that the
sun wheel as a whole changes its position in the heavens: in summer, it moves
toward the north along the axis of the heavens, and we see a large part of it
above the horizon, whereas in winter, we only observe a small part of the sun
wheel as it moves toward the south. This movement of the sun wheel ac-
counts for the seasons. The best way to describe this movement is that the
sun wheel as a whole slides regularly up and down a virtual cylinder, the
center of which is the celestial axis, the height of this cylinder being two
times that of the ecliptic, namely, 47°. Krafft, as shown in Figure 3.27, draws
FIGURE 3.27 The Right Picture of the Movement of the Sun Wheel
224 THE DISCOVERY OF SPACE
the first picture, known to me, which tries to visualize this idea.187 Unfortu-
nately, however, he rejects this representation in favor of an erroneous one, as
I just explained. The doxography tells us almost nothing about Anaximander’s
theory of the planets and their movements. In the only text in which they are
mentioned, they are bracketed with the stars.188 It is not difficult, however, to
explain their movements, using the same model, as holes in wheels that turn
around the earth from east to west, sometimes fastening and sometimes slow-
ing down their velocity, while moving more or less up and down, just like the
wheels of the sun and the moon.
I already have stated that I will take seriously the image of the celestial
bodies as “chariot wheels.” The supposition lies at hand that “the wheels of
the celestial bodies also had a height which is one-third their breadth, just
like the earth,”189 to use Rescher’s words. If we imagine such an Anaximandrian
celestial wheel, it will look like Figure 3.23, shown earlier. This version of the
celestial wheels looks remarkably like the pictures of chariot wheels, as was
shown in Figure 3.6. As we have seen, it is not too bold of a guess to say that
Anaximander has drawn a two-dimensional map of his conception of the
universe. We also have seen that the report that he made a three-dimensional
model in the shape of a sphere has to be considered an anachronism, and that
it is not likely that he made a three-dimensional model whatsoever. On the
other hand, I maintain that Anaximander’s conception of the universe is in
agreement with what can be observed in the heavens, or at least that it is not
a flagrant contradiction with the heavenly phenomena. Therefore, it must be
possible to draw a three-dimensional model that is both in agreement with
the doxographic tradition on Anaximander’s astronomy and explains the
movements of the celestial bodies as seen from a flat earth. Several scholars
have tried to construct a model of Anaximander’s universe, but without suc-
cess. The history of the visualization of the heavens according to Anaximander’s
ideas, is a concatenation of mistakes and misunderstandings, as I have shown
elsewhere. What will be represented below is an improved version of the
drawings I made earlier.190
When we accept that Anaximander had no knowledge of the ecliptic,
it is not too difficult to visualize his astronomical ideas. Anyone can see that
the sun travels on its daily path along a curve that makes a full circle, when
one imagines it going on at the other side of the earth. The wheel of the sun,
being as it were the materialization of this daily track, slides during the year
up and down an imaginary cylinder, the height of which is 47°, two times the
inclination of the ecliptic. This height can be measured with a gnomon with-
out knowledge of the ecliptic itself, as was shown in Figure 3.9. Inside of this
imaginary cylinder is another one for the moon. The movements of the moon
wheel resemble those of the sun, except that its rotation from east to west is
much slower, so the moon adopts its old position between the stars after
Dirk Couprie 225
approximately one month, and the virtual cylinder along which it slides up
and down is somewhat higher. As the monthly orbit of the moon makes an
angle of 5° with the ecliptic, the height of the moon’s virtual cylinder must
be 10° bigger than that of the sun. So the height of this cylinder amounts to
57°. Inside of these two virtual cylinders, we find the virtual sphere (or the
virtual cylinder, explained earlier) of the stars. At the center, the disk of the
earth is suspended. In this way, it is rather easy to explain the movements of
the celestial bodies by the movements of the celestial wheels.
Due to the inclination of the axis of the heavens, the celestial bodies do
not circle around the earth in the same plane as the earth’s—flat—surface.
The cylinders “lie aslant,” namely, with respect to the surface of the flat
earth.191 As Anaximander was not acquainted with the obliquity of the eclip-
tic (see the earlier discussion on the anachronistic fallacy), the expression
“lying aslant” (ke√menon lox¬n) in DK 12A22 “naturally refers to the more
obvious fact that the circles lie oblique to the plane of the earth’s surface,” as
West rightly remarks.192 To Anaximander, the tilting of the heavens’ axis, by
the way, must have appeared one of the biggest riddles of the universe. Why
is it tilted at all, who or what is responsible for this phenomenon, and why
is it tilted just the way it is? Unfortunately, the doxography has nothing to tell
us about this problem. Later, other Presocratics such as Empedocles, Diogenes,
and Anaxagoras mention the tilting (⁄gklisiV) of the heavens.193 This incli-
nation amounts to about 38.5° when measured at Delphi, the world’s navel.
The earth being flat, the inclination must be the same all over its surface. The
tilting of the heavens’ axis makes it possible that “the inner face of the wheel,
with the opening for the sun in it, could look down on the earth from above,”
according to O’Brien.194 It is the sun wheel’s height, being one-third of its
width, which we see from the earth, whereas it is the wheel’s width, and not
its height, which appears on the map of the universe. 195
In Figures 3.28a and 3.28b, the mechanism of the virtual cylinders is
shown.196 The sun wheel is rendered in both drawings at the top of its virtual
cylinder, that is, at the summer solstice. In Figure 3.28a, the sum total of the
star wheels is rendered as a virtual sphere, whereas it is drawn as a virtual
cylinder in Figure 3.28b, according to the alternative discussed above. The
wheels of the stars and those of the sun and the moon turn in the same
direction, from east to west, but with different speeds. The turning of the sun
wheel accounts for the succession of night and day, whereas the up and down
movement of the sun wheel along its virtual cylinder accounts for the chang-
ing of the seasons. In Figures 3.29a and 3.29b, the imaginary cylinders have
been omitted, and only the wheels of the sun and the moon are shown, on
different positions of their sliding movements along the virtual cylinders. For
clarity’s sake, two rather different positions have been represented: Figure
3.29a renders a possible situation on a night about the winter solstice (when
226 THE DISCOVERY OF SPACE
FIGURE 3.28A Anaximander’s Universe with the Virtual Cylinders of the Sun and Moon
Wheels and the Virtual Globe of the Star Wheels
FIGURE 3.28B Anaximander’s Universe with the Virtual Cylinders of all Celestial Wheels
(compare with Figure 3.26)
Dirk Couprie 227
Anaximander, looking up at the sky, studies the course of the celestial bodies),
whereas Figure 3.29b renders a possible situation on a day about the summer
solstice (when Anaximander, kneeling in the sand, measures the shadow of
his gnomon). In these pictures, four star wheels have been rendered (as cir-
cular lines), and the star wheels together are shown as making a virtual
sphere. This is mainly because in the case of the stars the image of the star
wheels taking the shape of a huge virtual cylinder, tempting as it may be, is
not necessary in order to show the movements of the stars, as seen from a flat
earth. In the case of the sun and moon wheels, however, the image of the
virtual cylinders is the only interpretation that fits both the observational
evidence (the movements of the sun and the moon as we see them) and the
doxographic tradition (of the celestial bodies as turning wheels).197
Il est difficile d’imaginer ce que ce peut être que ces trois figures de
femmes dans de si singulières attitudes.
The Egyptian word for “heaven” (“pt”) is written with the signs for “p” and
“t” over the ideogram for heaven. This ideogram is a simplified picture of a
flat roof (see Figure 3.30). The same picture is used many times in Egyptian
FIGURE 3.31 The Heavenly Roof on a Painting above the Entrance to Queen Nefertari’s
Burial Chamber
and Nut is the baboon-headed sky god, Shu, who holds up the heavens, his
raised arms making the “ka-”gesture as a symbol of cosmic power. In many
pictures, Nut not only bears stars all over her body but also is shown giving
birth to the sun (in the morning) and swallowing it again (in the evening).
Therefore, her head must be thought to be looking toward the west and her
rear toward the east.
When the French army, under Napoleon, conquered Egypt, another army
of scientists, which recorded its discoveries in the famous Description de l’Egypte,
followed it. One copied the ceiling of a room in the temple of Isis on the
island of Philae. This picture, which can be dated circa 140 B.C.E. (during the
reign of Ptolemaios VIII Eurgetes), exhibits three remarkable features. The
first is that we clearly see two Nuts, the one arching over the other, instead
of the traditional one goddess of the heavens.202 This representation breaks
with the common view of the one and only celestial vault by duplicating the
arching goddess of the heavens. The second feature is that on the body of the
FIGURE 3.34 Two Arching Nuts on the Ceiling of a Room in the Temple of Isis on the
Island of Philae
232 THE DISCOVERY OF SPACE
uppermost Nut, we recognize two winged suns, whereas on the belly of the
lowermost Nut, dotted stars can be seen. The most reliable copies count
twenty-four dots, twelve of them light and the other twelve dark (see Figure
3.34).203 The simplest explanation is that the stars are bright during twelve
hours of the day and not visible during twelve hours of the night (the hours
differing in length according to the season).204 The sun is on the upper Nut,
and the stars are on the lower one. This is the opposite of the order we would
expect, for in reality the sun is closer to the earth than are the stars. Other
duplicate representations of Nut exist, for example, on the famous roofs of the
tombs of Ramesses VI and Ramesses IX (see cover). The difference, however,
is that there the two Nuts are pictured in two clearly separated positions next
to each other, representing the daily heavens, with the birth and decline of
the sun, and the heavens at night, with the stars. In Figure 3.34, on the
contrary, the two Nuts are combined into one repetitive composition, the one
goddess arching over the other.
As a third feature we notice the strange position of the earth god, Geb.205
Ordinarily, Geb is pictured in a lying position, in a state of exhaustion. Here,
however, his body is stretched in a tense, acrobatic way. It has been explained
FIGURE 3.36 Acrobatic Osiris in the Book of Gates in the Osireion of Seti I (detail)
234 THE DISCOVERY OF SPACE
in Philae, another interpretation than those given earlier lies at hand. The
way in which Geb is rendered in Figure 3.34, as closely parallel to Nut as
possible, must be intentional. The strange bow of Geb’s body looks like a
repetition of the two arched bodies of Nut. In other words, it looks as if the
artist has meant to hint at a threefold representation of the heavens. This
impression is particularly strong when the right half of the picture is covered.
This interpretation is supported by the androgynous way in which Geb’s body
is represented. The suggestion is so strong that it has deluded at least one
scholar, as we will see soon.
What is the explanation of this picture of a twofold Nut, and what is the
meaning of the reversed order of the stars and the sun on her two bodies?
Why this sudden deviation of an age-old tradition of representing the heav-
ens? Before we try to answer these questions, let us look at another picture.
Victor Denon, another scientist in the train of Napoleon, was the first to
make a copy of it (see Figure 3.37).212. We see three goddesses of the heavens
arching over one another. This time, Geb is missing. The picture is from the
temple of Hathor in Dendara, on the ceiling of the room next to that where
the famous map of the heavens, known as the “Zodiac of Dendara,” which
is now in the Louvre Museum in Paris, has been found. On this Zodiac,
Greek influence is unmistakable, for not only traditional Egyptian constella-
tions are depicted but also the Greek zodiacal signs. The Zodiac has recently
been dated to June–July 50 B.C.E., as could be deduced from the special
constellation of planets in the signs of the zodiac, as well as from the location
of a solar and lunar eclipse.213 The picture of the three Nuts presumably dates
from the same time. Daumas mentions this picture of “trois déesses emboîtées,
si l’on peut dire, les unes dans les autres.”214 Discussing Lanzone’s illustration
of the two Nuts and Geb in Philae, he says: “malheureusement (. . .) certaines
de ces planches sont fausses (CLV, 2e figure il manque une des trois Nout
emboîtées, signalées ici, note 1, p. 1) (. . .).”215 So he takes Lanzone’s drawing
of the two Nuts in Philae as an unsuccessful copy of the ceiling with the three
Nuts in Dendara. This mistake is instructive, as it supports my impression
that the Egyptian artist of the ceiling in Philae intentionally produced the
suggestion of a triplicate Nut.
All of a sudden, at the time of the Ptolemaic dynasty, the representation
of the heavens by one arched Nut, which has been identical for many ages,
is replaced by a twofold or threefold Nut. So far, nobody has given a satisfying
interpretation of the representations of the two Nuts in Philae and the three
Nuts in Dendara. With reference to the ceiling with the two Nuts, Schäfer
remarks that it proves that the Egyptian mythologists supposed the existence
of more than one heaven.216 This statement contradicts the fact that until
Ptolemaic times, the Egyptians used to depict the heavens by only one arch-
ing Nut. Concerning the ceiling with the three Nuts, Denon wonders what,
Dirk Couprie 235
FIGURE 3.37 Three Goddesses of the Heaven Arching over One Another on the Ceiling of
a Room in the Temple of Hathor in Dendara
for heaven’s sake, these three figures may mean.217 Cauville suggests that the
three Nuts of Dendara have the same meaning as the three boxes of
Tutankhamun’s sarcophagus.218 I regard this as sheer fantasy.
Cauville makes a more serious remark when she compares this picture of
the three Nuts to the two Nuts in the temple of Isis in Philae.219 Reflecting
on the picture of the two Nuts in Philae and the three Nuts in Dendara, I
think that two features of these representations of the heavens point in a
certain direction. The first feature is the strange order of the celestial bodies
in the first one, with the stars closer to the earth than the sun; the second is
the threefold curvature of the heavens in the second one, and perhaps also in
the first. Both features remind me of Anaximander. His astronomical concep-
tion results in a picture similar to Figure 3.21. The images are different,
236 THE DISCOVERY OF SPACE
turning chariot wheels versus arching goddesses, but two main features are
exactly the same, namely, the threefold stratification of the heavens, and the
upside-down order of the celestial bodies. The pictures of the double and
triple Nuts look as if the artists have tried to reconcile this revolutionary
astronomical insight with the traditional way of depicting the heavens. My
hypothesis is that it must have been the influence of Anaximander’s teachings
that made the Ptolemaic Egyptians introduce such a drastic change in a
representation of the heavens that has been the same for many centuries.
Perhaps three questions have to be answered before the reader is inclined
to accept this interpretation. The first is how could Anaximander’s conception
of the universe have influenced Egyptian representations so far away from his
birthplace? The second is why has it taken so long before it exercised its
influence? And the third is why, of all the Greek astronomers and philoso-
phers, was it Anaximander the Egyptians took the pains to deal with? First,
we must not underestimate the measure of exchange of philosophical and
scientific ideas between the countries around the Mediterranean Sea. On the
other hand, the rate of exchange of ideas was necessarily much slower than
what we are used to. A factor of importance is that since the time of Alexander
the Great, Greek (Ptolemaic) dynasties ruled over Egypt. The Greeks were
used to looking with awe at the Egyptian civilization. In the days from Thales
to Plato, the idea was that one had to go to Egypt in order to get real
wisdom. Since Alexander had conquered the land and had become ruler over
Egypt, the tables were turned, and gradually the Greek influence on the
Egyptian culture was substantial. In the famous library of Alexandria, all of
the wisdom of that time was stored. It was founded in the third century B.C.E.
by Ptolemy I. Anaximander’s book seems to have been in this library, where
Apollodorus (second century B.C.E.) should have consulted it.220 The ceilings
with the double and triple Nuts were created in the second and first century
B.C.E. in temples deep in the southern part of Egypt, about 800 and 1,000 km
from Alexandria. This quick survey of some relevant facts allows us to under-
stand why it was Egypt where Anaximander’s ideas became known, and why
it took so long before they arrived at Philae and Dendara.
Regarding the third question, it is appropriate to remark that Greek
astronomical ideas gradually influenced the Egyptian astronomy. As an ex-
ample, I already have mentioned the Greek zodiacal signs and other constel-
lations, next to the traditional ones, on the Zodiac of Dendara. Anaximander’s
conception of the universe, however, when it had reached Egypt, must have
meant a special challenge to the Egyptian astronomers. It was precisely his
main idea of what I call “depth in the universe,” that broke with the age-old
tradition of the representation of the celestial vault. The Egyptian astrono-
mers also must have observed that the idea of a stratified universe had be-
come a constituent of Greek astronomic thinking as such. Therefore, we may
Dirk Couprie 237
imagine that the Egyptian astronomers must have been anxious to incorpo-
rate this idea that seemed so threatening to their own opinions. Their chal-
lenge was, so to speak, to show that this new idea was somehow compatible
with the age-old representations of the celestial vault by an arching Nut. This
is exactly what we see on the ceilings of the temples in Philae and Dendara.
Obviously the Ptolemaic astronomers tried to incorporate Anaximander’s
revolutionary view of the universe into the traditional Egyptian representa-
tions of the heavens, and they told the artists how to achieve this goal.
CONCLUSION
82. Heidel (1937, 12 n.): “It is quite certain that the continental mass, not to
speak of the oikumene, was not circular, though the map probably was in the earlier
times.” Agathemerus (DK 68B15 (2) [16]) relates that the Ancients described the
inhabited earth (oÎkoum°nh) as circular, whereas Democritus was the first to conceive
of the earth (g›) as oblong. This remark would make more sense if the words
oÎkoum°nh and g› were reversed.
83. Hahn (2001, 205–98). See also Couprie and Pott (2002) for an extensive
review of Hahn’s book. Heidel (1937, 58), mentions “the undoubted relation of the
Ionian map to the horizon and the points that were marked on the dial of the
gnomon.”
84. Heidel (1937, 18 ff.).
85. Ibid., 20.
86. Ibid., 54.
87. Brumbaugh (1981, 22); Lee (in Aristotle, 1952, between 102 and 103), draws
the “equinoctial line” exactly through these three locations. See also Heidel (1937, 6,
Fig. 1). Herrmann (1931, 47, Abb. 3), also draws an “equator” through Delphi and the
Pillars of Hercules, but somewhat north of Miletus.
88. See Heidel (1937, 32).
89. Herodotus, op. cit., loc. cit.
90. Heidel (1937, 12, cf. 22 and 31). This also is the way that Hecataeus’ (a successor
of Anaximander’s) map is rendered in Bunbury (1879, I, between 140 and 141) and in
Bengtson (1963, 8, Fig. c). See also Berger (1903, 78), who remarks that the majority of
the sources ascribes to the ancient geographers a division of the earth in two parts, and
then continues at 81: “Wenn er (Herodotus) aber die Jonier darum tadelt, daß sie Europa
und Asien gleich machen, dann kann er sich nur die kreisförmige Ökumene in einen
nördlichen und südlichen Halbkreis zerlegt vorgestellt haben.” Elsewhere, however,
Herodotus says that the Ionians distinguished three parts of the world: Europe, Asia, and
Libya, the river Phasis being the boundary between Europe and Asia, and the river Nile
between Asia and Libya, whereas Europe is represented as big as Asia and Libya together.
(Herodotus, Histories, II, 16). Robinson’s reconstruction (1968, 32) divides the surface of
the earth in three parts. See also Hahn (2001, 210), and Naddaf, in this volume.
91. In fact, it is no more than a suggestion, made by Mieli (1916, 63n.). See also
Conche (1991, 46).
92. See DK 12A2 [10], 12A6 [27], DK12A6 [30], 12A1 (2) [16]).
93. From Kahn (1960/1994, between 88 and 89), (British Museum 92687).
94. Kahn (1960/1994, 83n.).
95. Line drawing from Clère (1958, 30). A photograph of the sarcophagus that
is in the Metropolitan Museum of Arts, New York (BMMA 9,117), in Betró (1999,
156). She dates it at the twentieth dynasty, but this is probably a misprint; Owusu
(1998, 15) dates it at the fourth millennium B.C.E., which is nonsensical.
246 THE DISCOVERY OF SPACE
121. DK 67B2 [5]. This text, by the way, shows that the principle was known to
the Presocratics. Therefore, Heidegger is exaggerating when he says that it has taken
2,000 years since the dawn of Western philosophy, before the “Satz vom Grunde” was
formulated. See Heidegger (1957, 14–15).
122. Barnes (1982/1979, 25).
123. Von Fritz (1971, 24).
124. I know of only a few exceptions: (1) Sticker (1967, 20): “Was veranlaßte
wohl diese erstaunliche Hypothese? (viz., daß nämlich die Erdscheibe frei im Raume
schwebt). Vielleicht die vernünftige Überlegung, daß nämlich von allen anderen
denkbaren Möglichkeiten diese allein es zuläßt, sich vorzustellen, daß die Gestirne im
Laufe von Tag und Nacht eine volle Kreisbahn—nicht nur immer wieder einen neuen
Tagesbogen von Aufgang bis Untergang—um die in ihrer Mitte ruhend gedachte
Erde volführen können.” (2) Von Fritz (1971, 143): “Aber die Vorstellung von den
Gestirnen als Ringen, die sich um die ganze Erde herumlegen, hat es ihm ermöglicht,
einzusehen (. . .) daß die Richtungen nach oben und unten nicht (. . .) absolute sind
(. . .), womit sich dann auch das weitere alte Problem erledigt, worauf die Erde eigentlich
ruht oder warum sie nicht ins Bodenlose fällt.” See also Furley (1987, 28).
125. As will be explained herein after, there is one other movement that is
allowed to these celestial rings: up and down along the axis of the heavens. But also
during this movement, the distance from the earth to any point at the circumference
of the ring is equal at any given time.
126. See Dicks (1970, 45–46).
127. Iliad, XVII, 425; Odyssey, I, 52–54; III, 2; XV, 329; XVII, 565. Cf. Dicks
(1970, 30).
128. See Dreyer (1953, 7); Pindar (1997, vol. 1, 360, line 27): ˛ cºlkeoV o¶ran¿V;
Theognis (1979, 146, lines 869–70): m°aV o¶ran¿V e¶r∞V ©perJen, cºlkeoV.
129. Burnet (1945, 69).
130. DK 13A14 [26].
131. Eisler (1910, 90–91, n. 3). See also Bousset (1960, esp. 24ff.). Later scholars
point out similar parallels. See Burkert (1963, 106, 110–12), and West (1971, 89–90).
132. Most recently by Schmitz (1988, 77–78). See also Kahn (1960/1994, 90 n.
1), and Duchesne-Guillemin (1966, 425).
133. This concerns the so-called Hilprecht text, HS 229. See Neugebauer (1952,
94, part of the quoted text), and Rochberg-Halton (1983, 212). AN.TA.GUB is an,
as yet, unidentified star. I thank Dr. W. H. Van Soldt from the University of Leiden
for his kind information on this issue.
134. Rochberg-Halton (1983, 216). See also Van der Waerden (1974, 62ff.).
135. Tablet VAT 8917, lines 30–33. See Livingstone (1989, 100); see also
Livingstone (1986, 83), Lambert (1975, 58), and West (1997, 139), who dates the
tablet in the mid seventh century B.C.E. Kingsley (1992, 341–430) compares this text
248 THE DISCOVERY OF SPACE
with Ezekiel 1: 26–7, where God is sitting in his place above the firmament, just like
Bel is seated on his throne above the heaven of the stars.
136. Heidel (1937, 7).
137. See DK 12A10 [33].
138. In DK 12A10 [36], the word ™porrage√siV (“violently tearing free”) is
used.
139. So already Diels (1897, 229). See also Freudenthal (1986, 212).
140. Krafft (1971a, 106).
141. See, for example, Kahn (1906/1994, 91); Rescher (1958, 724, Fig. 5). Repr.
in Rescher (1969, 3–22).
142. Eisler (1910, 90n.); Diels (1897, 233); Burkert (1963, 110–11); West (1971,
90–91).
143. West (1971, 92).
144. DK 12A11 [1].
145. Bodnár (1988, 50).
146. DK 67A1 (33) [15].
147. Cf. Aristotle, who argues that the celestial bodies cannot be at an infinite
distance, because in that case they would have to circle around the earth with an
infinite velocity (De Caelo, 272b19 ff.). This argument, however, is fallacious, because
the center of an infinite universe is everywhere, as Giordano Bruno later acknowl-
edges.
148. DK 12A11 (5), a mutilated text, corrected by Diels; idem DK 12A21 [15].
149. DK 12A21 [10].
150. DK 12A22 [18].
151. Cf. DK 12A18 [31]: •p¿ dÆ a¶to∞V (viz., sun and moon) tΩ ™plan› tÍn
Òstrwn ka¥ to∞V plºnhtaV.
152. Tannery (1887/1930, 94–95); Diels (1897, 236); O’Brien (1967, 423–27).
153. Kirk, Raven, and Schofield (1995, 136, n. 1).
\ lion ’√son eΔnai t˛› g˛›.
154. DK 12A21 [14]: 'Anax√mandroV t¿n m°n ≠
155. As O’Brien himself remarks (1967, 424n.), this was in fact the intention of
Tannery and Diels. Tannery says explicitly: “la double épaisseur du cerceau est ainsi
égale au diamètre de la Terre” (1887/1930, 94). And Diels says: “so ist die Breite dieser
Ringe auf einen Erdradius zu veranschlagen” (1897, 232). Diels’ drawing also leaves
no doubt about his intentions.
156. O’Brien (1967, 423–24).
157. DK 12A1 (2) [16], 12A2 [10].
158. DK 12A2 [9].
Dirk Couprie 249
159. This picture is essentially the same as Conche’s, who, however, draws the
disk of the earth some three times too big. See Conche (1991, 210).
160. O’Brien (1967, 427). Engmann’s opinion that “the figures do not yield equal
distances” is clearly wrong (Engmann 1991, 22).
161. DK 12A17 [19]. See also Cornford (1934, esp. 12). Cf. also Kahn (1960/
1994, 46–53, esp. 50).
162. Kahn (1960/1994, 96, emphasis added).
163. Dicks (1966, 36). Conche’s conjecture that Anaximander put the sun at such
a distance that it appeared as a disc and not as a point does not lead to any numbers
at all, notwithstanding what he seems to imply (1991, 218–19).
164. Kahn (1960/1994, 80). Elsewhere, however, he says that “the celestial di-
mensions given by Anaximander cannot have been based upon any kind of (. . .)
geometric construction” (op. cit., 94).
165. Hahn (2001, 163). For a critical review of Hahn’s attempt to associate
Anaximander’s alleged mathematics with the methods of ancient architects, see Couprie
and Pott (2002).
166. Aristarchus of Samos (310–230 B.C.E.) was the first who seriously tried to
measure the (relative) distances of the sun and the moon. The problem of measuring
the sun’s distance, and thereby its actual size, however, was beyond the instrumental
capacity of astronomers until the invention of the telescope, and even now the mea-
surement of distances in the universe is one of the biggest problems in astronomy.
167. k√oni l√Jw˛ parapl–sion, DK 12A11 (3) [8]; cf. 12A25. Perhaps the
shape of the earth also has to do with the image of the wheels, for it might be seen
as the hub of the concentric celestial wheels, especially when drawn on a two-dimen-
sional map.
168. Bayhan (1993, 19).
169. West (1971, 94); Krafft (1971a, 107). See also Burkert (1962, 288, n. 63).
170. Another example of the number 9 indicating a very long time is in Hesiod’s
Theogony (775–806), where a god who inadvertently drinks from the river Styx is said
to be exiled for nine full years from the everlasting gods and to be allowed to return
in the tenth year.
171. Krafft (1971a, 107–108). See also Germain (1954, 13).
172. In Couprie (1995, 162), I have followed the usual view, which has the
numbers refer to the diameters of the wheels of the celestial bodies. This, however,
does not invalidate the conclusions of that article.
173. I owe the idea of putting suns one after another on the circumference of the
circle of the sun ring, in order to visualize the angular diameter of the sun, to Stritzinger
(1952, 63).
174. See Figure 2.6 in Couprie (2001a, 42). This problem is even bigger if we
look at the numbers as indicating diameters of the rings. In that case, we get 81 suns
250 THE DISCOVERY OF SPACE
(2 (= 3) 27), put one after another to make up the total sun ring. In other words,
the angular diameter of the sun would in that case amount to approximately 4°30',
which is about nine times too big, the angular diameter of the sun being, in fact, 30'.
175. Here I strongly disagree with Dreyer, who says that “no doubt this might
have escaped his attention” (1953, 15n.).
176. Not 1/27, as Hölscher maintains (1970, 318n.). See DK 11A1 (24), 11A19.
Diogenes Laertius’ account, however, seems not to be consistent with what we know
about Thales’ cosmology. Cf. the critical notes in Dicks (1959, 306; 1966, 27).
177. DK 12A18: pil–mata ™°roV trocoeid›, pur¿V ⁄mplea (Kahn’s [1994,
86] translation).
178. o≥on prhst›roV a¶l¬n (DK 12A22). For this translation of what, since
Diels, is called “the nozzle of a bellows,” see Couprie (2001b).
179. Tannery (1887/1930, 97).
180. Krafft (1971a, 116, Abb. 5, left). For a more thorough investigation of
wrong pictures of Anaximander’s universe, see Couprie (1995, 159–81).
181. Sarton (1959, 175, emphasis added).
182. Heath (1921, 35–36); Neuhäuser (1883, 408, 427); Couprie (1995, esp.
169–73).
183. All of these authors seem to confuse the plane of the horizon with that of
the celestial equator, for it is true that the ecliptic intersects the equator at two definite
points, the northern vernal equinox and the northern autumnal equinox.
184. Mulder (1979, 55–59, and the drawings on 58–59).
185. The same idea is in Kahn (1970, 107): “Now the wheels of the fixed stars
are presumably oriented so as to account for the diurnal motion. That means that they
are, in effect, set perpendicular to the axis of rotation. If these circles are all assumed
to have the same diameter, the result will be a cylinder.” Kahn rejects this possibility,
because “it would in any case be incompatible with Anaximander’s conception of
celestial symmetry, since the cylinder would have to be set obliquely to the plane of
the earth.” This is no strong argument, as the wheels of the stars also are set obliquely
to the plane of the earth.
186. Photograph by Remco Scheepmaker.
187. Krafft (1971a, 116, Abb. 5 right). The same idea has been expressed in other
words by Szabó when he describes how one sees the movements of the sun from the
earth: “Aber noch einfacher denkt man sich die täglichen Bogen der Sonnenlaufbahn
am Himmel—wenn der Sommer herannaht—als immer größer werdende Abscnitte
von parallelen Kreisen.” (Szabó and Maula 1982, 37).
188. DK 12A18 [31].
189. Rescher (1958, 727). In his pictures, however, Rescher does not stick to his
own prescription (op. cit., Figs. 9 and 10).
Dirk Couprie 251
a cutout wraparound that easily can be taped to a twelve-ounce beer or soft drink can
(or a Campbell’s soup can). Actually, Gingerich uses his model of what he calls “the
cola-can universe” to illustrate the dance of the ecliptic, and that is why his cylinder
turns the other way around than mine, but the principle is the same.
198. Drawing by Hans Exterkate after Clayton (1994, 148).
199. Drawing by Hans Exterkate after Leclant (1981, 119), seventh century
B.C.E.(?) The date of this picture is uncertain. Silverman (1997, 170), gives as the date
c. 1000 B.C.E., and Schulz and Seidel (1997, 432) as c. 850 B.C.E.
200. Drawing by Hans Exterkate after Leclant (1980, 120).
201. See Note 2.
202. See Pancoucke (1820, A., vol. 1, pl. 10 (1); see also A. Vol. 1, pl. 18). It has
been described in Pancoucke (1821, Tome Premier, 62) as: “Ce bas-relief (. . .) est
d’une grande singularité par l’enroulement, on peut dire monstrueux, des trois figures
qui le composent.” In Pancoucke (1820, A., vol. 2, pl. 37), the same picture appears
three times on the representation of the hindmost ceiling of a temple of Hathor
(wrongly called the temple of Isis) at Deir-el-Medina. However, these are additions
of the French artist, probably for aesthetic reasons. Porter and Moss (1972, 401–407)
describe this temple, but do not mention any representations on the ceilings.
203. Bénédite (18, 93, pl. L); see also Denon (1990/1803, pl. CXXIX, no. 5);
Prinz (1915, Tafel VII, 2); Lanzone (1974, vol. 1, TAV. CLV [b]). There exists a (very
vague) photograph of this ceiling in a publication that is only available on microfiche
(see Junker and Schäfer 1975, photo 1246).
204. This invalidates an interpretation, mentioned by Prinz (1915, 22), in which
the lower Nut should be the night sky and the upper Nut the heaven at day. For
similar reasons, these twenty-four dots cannot be explained as the sun disks that are
sometimes seen on pictures of Nut, representing the nightly passage of the sun through
her body.
205. Another example of Geb in this strange position is in the temple of Hathor
in Dendara. See Lamy (1981, 21). For the first reproduction of this painting, see
Denon (1990/1803, pl. CXXIX, no. 8).
206. Author’s drawing, after Omlin (1973, pl. XXVIII[b]), see also Lanzone
(1881–86/1974, Vol. I, TAV. CLIX), and Parkinson (1999, 170).
207. See Niwiński (1989, 199). This qualification, however, holds true especially
for the physically impossible posture of Geb’s legs, with the knees bending the wrong
way. Kaper’s (1997, 143 n. 13) suggestion that Nut has been pictured as a male god
“because she is already present in the adjacent vignette on the same papyrus and
repetition was not deemed desirable,” is not satisfying. Parkinson’s (1999, 170)
identification of “an ithyphallic figure of Osiris representing the night sky,” given
without any explanation, is certainly wrong.
208. Omlin (1973) does not discuss this picture, but the fact that he includes it
in his book implies that he considers it as satyric-erotical.
Dirk Couprie 253
209. See Budge (1913, 134); see also Naydler (1996, 27).
210. This is the interpretation of a similarly curved figure Hathor’s temple at
Dendara in Cauville (1997c, 48).
211. I thank my friend, Ton Verschoor, for drawing my attention to this differ-
ence in attitudes.
212. Cauville (1997a, 260, and 283, photo); see also: Cauville (1997c, 75), and
Denon (1803, pl. CXXIX, no. 6). Although Pancoucke (1820) does not have a copy
of it, it is described in Pancoucke (1821, Tome Troisième, 369): “(. . .) trois femmes
emboîtées, pour ainsi dire, les unes dans les autres (. . .) elles sont surtout dignes
d’attention, à cause de la disproportion choquante de tous leurs membres (. . .).”
213. See Aubourg (1995). Cauville (1997c, 76–77), gives the date December 28,
47 B.C.E., but this difference does not bother us here.
214. Daumas (1951, 373n).
215. Ibid., 375n.
216. Schäfer (1928, 107).
217. Denon (1990/1803, pl. CXXIX, no. 6). The same astonishment is expressed
in Pancoucke (1821, 369–70): “(. . .) l’on ne peut douter qu’elles ne soient des êtres de
convention pour exprimer de certaines choses dont nous ne pouvons plus maintenant
deviner le sens.”
218. Cauville (1997c, 75).
219. Cauville (1997b, 204).
220. See Diels (1879, 219, n. 3). See also Heidel (1921, 261): “It was there in
all probability that Apollodorus found the book.” Recently evidence has appeared that
Anaximander’s book must have been available in the second century B.C.E. in Taormina
in Sicily, where a fragment of a catalogue of a library has been found, on which the
words 'Anax√mandroV Praxiºdou Mil–sioV and ÷g°[n]eto mÆn Q[al]°w can be
read (see Blanck 1997, 509).
221. Furley (1987, 29); Kahn (1960/1994, 233).
222. Furley (1987, 30).
223. Cornford (1934, 2).
224. McKirahan (2001, 54).
225. Cornford (1934, 10).
226. See DK 13A1, 13A7 (6), 13A14; DK 13A6, 13A7 (4), 13A20; DK 13A13,
13A14, 13A15.
227. See DK 13A1, 12A9, 12A8.
228. Johnson (1937, 166). Digges’ text is reproduced in Johnson and Larkey
(1931, 79–95).
229. Furley (1987, 1–2, and passim).
254 THE DISCOVERY OF SPACE
230. This is the more fundamental reason I agree with McKirahan (2001, 65)
that Anaximander cannot be placed in either the Closed World or the Infinite Uni-
verse camp, although I do not think that Anaximander believed in an infinite number
of worlds.
231. After having finished this essay on Anaximander’s astronomy, I feel obliged
to express my thanks to those who have been of help in one way or another. First of
all, of course, I have to thank my co-authors Robert Hahn and Gerard Naddaf, with
whom I have shared in countless e-mails and in too few actual conversations almost
every twist of my ideas. Without their encouraging criticism I would have given up
hope to understand Anaximander’s astronomy altogether already years ago. At crucial
points in the development of my thoughts, István Bodnár answered my e-mails with
his profound scholarly remarks. With Heleen Pott I discussed almost every word of
the manuscript. Her professional support as a philosopher has shaped my understand-
ing of Anaximander in a decisive way. My friend Ton Verschoor has read an early
version of the manuscript as a non-professional. His painstaking criticism has made
me think many things over again and has led to many improvements. My friend Hans
Exterkate was of invaluable help in drawing a number of beautiful illustrations and
thus in overcoming the silly difficulties of copy-rights. Remco Scheepmaker kindly
permitted the use of his photograph of the stars circling around the Polar Star. Dr. W.
H. van Soldt, Dr. J. F. Borghouts, and other members of the NINO (Dutch Institute
of Near Eastern Studies) of Leiden University were always very kind and patient in
answering my questions on several Babylonian/Assyrian and Egyptian topics.
Bibliography 255
Bibliography
255
256 BIBLIOGRAPHY
———. 1945 (4th. ed.). Early Greek Philosophy. London: Adam and Charles Black.
Burton, A. 1972. Diodorus Siculus. Book I: A Commentary. Leiden: Brill.
Buschor, E. 1930. “Heraion von Samos: Frühe Bauten.” Athenische Mitteilungen 55: 1–99.
Cauville, S. 1997a. Dendara X, Les chapelles osiriennes. Le Caire: Institut Français
d’Archéologie Orientale.
———. 1997b. Le temple de Dendara. Les chapelles osiriennes. 2, Commentaire. Le Caire:
Institut Français d’Archéologie Orientale.
———. 1997c. Le Zodiaque d’Osiris. Leuven: Peeters.
Cherniss, H. 1951. “The Characteristics and Affects of Presocratic Philosophy.” Jour-
nal of the History of Ideas 13(3): 319–45. Reprinted in pp. 1–28 of Studies in Presocratic
Philosophy, Vol. I: The Beginnings of Philosophy, eds. D. J. Furley and R. E. Allen.
1970, New York and London: Humanities Press.
Classen, C. J. 1966. “k√wn l√JoV.” The Classical Review 16: 275–76.
———. 1986. Ansätze. Beiträge zum Verständnis der frühgriechischen Philosophie.
Würzburg and Amsterdam: Rodopi.
Clay, D. 1992. “The World of Hesiod.” Pp. 131–55 in. Essays on Hesiod II, edited by
A. N. Athanassakis, Bendigo North: Aureal Publ.
Clayton, P. A. 1994. Chronicle of the Pharaohs. London: Thames and Hudson.
Clère, J. J. 1958. “Fragments d’une Nouvelle Représentation Égyptienne du Monde.”
Mitteilungen des deutschen archäologischen Instituts, Abteilung Kairo 16: 30–46.
Coldstream, J. N. 1977. Geometric Greece, New York: St.Martin’s Press.
Cole, I. H. 1926. “Determination of the exact size and orientation of the Great
Pyramid at Giza,” Archeological Survey of Egypt, Paper No. 39.
Colemen, J. E. 1996 “Did Egypt Shape the Glory That Was Greece?.” Pp. 280–302
in Black Athena Revisited, eds. M. Lefkowitz and G. McLean Rogers. Chapel Hill,
N. C.: University of North Carolina Press.
Conche, M. 1991. Anaximandre. Fragments et Témoignages. Paris: Presses Universitaires
de France.
Cornford, F. M. 1934. “Innumerable Worlds in Presocratic Philosophy.” The Classical
Quarterly 28: 1–16.
Coulton. J. J. 1975. “Towards Understanding Greek Temple Design: General Consid-
erations,” The Annual of the British School at Athens 70: 59–99.
———. 1976. “The Meaning of 'Anagraje§V”, American Journal of Archeology 80:
302–4.
———. 1977. Ancient Greek Architects at Work. Ithaca, N. Y.: Cornell University Press.
Couprie, D. L. 1995. “The Visualization of Anaximander’s Astronomy.” Apeiron 28:
159–81.
———. 1999. “’Hätte die Welt ein Ziel, [. . .] so wäre es [. . .] mit allem Werden
längst zu Ende.’ Ein Beitrag zur Geschichte einer Argumentation,” Nietzsche-Studien
27 (1998): 107–18.
258 BIBLIOGRAPHY
Dillon, M., and L.Garland, 1994. Ancient Greece: Social and Historical Documents from
Archaic Times to the Death of Socrates. London: Routledge.
Dinsmoor, W.B. 1975 The Architecture of Ancient Greece: an Account of Its Historic
Development. New York: W. W. Norton. (Reprint of 1902. London: no publisher
indicated).
Diodorus of Sicily, 1970–1989. Works. 12 vols. with English translation by C. H.
Oldfather et. al. Cambridge, Mass.: Harvard University Press.
Drerup, H. 1969. Griechische Baukunst in geometrischer Zeit, Archeologica Homerica.
Göttingen: Vandenhoeck und Rupprecht.
Dreyer, J. L. E. 1953. A History of Astronomy from Thales to Kepler. New York: Dover
Publications. Facs. reprint of 1905 History of the Planetary Systems from Thales to
Kepler. Cambridge: Cambridge University Press.
Duchesne-Guillemin, J. 1966. “D’Anaximandre à Empédocle: contacts Gréco-Iraniens.”
Pp. 423–31 in Atti del convegno sul tema: la Persia a il mondo Greco-Romano (Roma,
11–14 aprile 1965), ed. A. Pagliano. Roma: Accademia Nazionale di Lincei.
Dumont, J.-P. 1988. Les présocratiques. Paris: Gallimard.
Ebeling, E. 1931. Tod und Leben nach der Vorstellung der Babylonier. I.Teil, Texte. Berlin
und Leipzig: Walter de Gruyter & Co.
Eggermont, P. H. L. 1973. “The Proportions of Anaximander’s Celestial Globe and
the Gold-Silver Ratio of Croesus’ Coinage.” Pp. 118–28 in Symbolae Biblicae et
Mesopotamicae Francisco Mario Theodoro De Liagre Böhl Dedicatae, eds. M. A. Beek
et al., Leiden: E. J. Brill.
Eisler, R. 1910. Weltenmantel und Himmelszelt. München: Beck.
Engmann, J. 1991. “Cosmic Justice in Anaximander.” Phronesis 36(1): 1–25.
Eubulus. 1952. The Fragments. Edited by R. L. Hunter. Cambridge: Cambridge
University Press.
Fehling, D. 1985. “Das Problem der Geschichte des griechischen Weltmodells vor
Aristoteles.” Rheinisches Museum für Philologie 128: 195–231.
Forrest, G. 1986. “Greece: The History of the Archaic Period.” Pp.13–43 in Greece and
the Hellenistic World, eds. J.Boardman, J. Griffin and O. Murray. Oxford and New
York: Oxford University Press.
Fränkel, H. 1973. Early Greek Poetry and Philosophy. New York and London: Harcourt
Brace Jovanovich. Translation of Dichtung und Philosophie des frühen Griechentums.
München: Beck 1962, by M. Hadas and J. Willis.
Frank, E. 1962. Plato und die sogenannten Pythagoreer. Ein Kapitel aus der Geschichte des
griechischen Geistes, Darmstadt: Wissenschaftliche Buchgesellschaft. Reprint of 1923,
Halle: Niemeyer.
Frankfort H., H. A. Frankfort, J. A.Wilson, and T. Jacobsen, 1946. Before Philosophy.
Harmondsworth, Middlesex: Penguin Books Ltd.
Freudenthal, G. 1986. “The Theory of the Opposites and an Ordered Universe: Physics
and Metaphysics of Anaximander.” Phronesis 31: 197–228.
260 BIBLIOGRAPHY
———. 1992b. “Anaximander and the Architects.” Proceedings of the Society for Ancient
Greek Philosophy: 1–26.
———. 1995. “Technology and Anaximander’s Imagination: A Case-Study for the
Influence of Monumental Architecture on the Origins of Western Philosophy/
Science.” Pp. 93–136 in New Directions in the Philosophy of Technology. ed. J. C. Pitt.
Dordrecht, Boston, and London: Kluwer Academic Publishers.
———. 2001. Anaximander and the Architects. The Contributions of Egyptian and Greek
Architectural Technologies on the Origins of Greek Philosophy. Albany: State University
of New York Press.
Hall, E. 1996. “When is Myth Not a Myth?: Bernal’s ‘Ancient Model.’” Pp. 333–48
in Black Athena Revisited, eds. M. Lefkowitz and G. McLean Rogers, Chapel Hill,
N.C.: The University of North Carolina Press.
Haselberger, L. 1985. “The Construction Plans for the Temple of Apollo at Didyma.”
Scientific American 253(6), (December): 126–32.
Havelock, E. A. 1957. The Liberal Temper in Greek Politics. New Haven, Conn.: Yale
University Press.
———. 1978. The Greek Concept of Justice. From Its Shadow in Homer to Its Substance
in Plato. Cambridge: Cambridge University Press.
Heath, T. L. 1908. The Thirteen Books of Euclid’s Elements. Cambridge: Cambridge
University Press.
———. 1921. A History of Greek Mathematics, Volume I, From Thales to Euclid. Oxford:
Oxford University Press.
Heidegger, M. 1975a. Der Satz vom Grund. Tübingen: Max Niemeyer Verlag.
———. 1975b. Early Greek Philosophy. Translated by F. A. Capuzzi. New York ans
Evanston: Harper and Row.
Heidel, W. A. 1906. “The D√nh in Anaximenes and Anaximander.” Classical Philology
1: 279–82.
———. 1921. “Anaximander’s Book: The Earliest Known Geographical Treatise.”
Proceedings of the American Academy of the Arts and Sciences 56: 239–88.
———. 1935. Hecateus and the Egyptian Priests in Herodotus, Book II. New York:
Garland Pub.
———. 1937. The Frame of the Ancient Greek Maps. With a Discussion of the Discovery
of the Sphericity of the Earth. New York: American Geographical Society.
———. 1943. “Hecataeus and Xenophanes.” American Journal of Philology, 64: 257–77.
Herodotus. 1921–1924. Histories. Edited and translated by A. D. Godley. London and
New York: Heinemann.
———. 1987. The History. Translated by D. Grene, Chicago: University of Chicago Press.
262 BIBLIOGRAPHY
Kahn, C. H. 1970 “On Early Greek Astronomy.” The Journal of Hellenic Studies 90:
99–116.
———.1994. Anaximander and the Origins of Greek Cosmology. Indianapolis and Cam-
bridge: Hachett Publishing Company Ltd. Reprint of 1960, New York: Columbia
University Press.
Kambartel, F. 1966. “Der Satz vom zureichenden Grunde und das Begründungsproblem
der Mechanik. Zu einer Bemerkung von Leibniz im 2.Schreiben an Clarke.”
Zeitschrift für philosophische Forschung 20: 457–70.
Kaper, O. E. 1997. Temples and Gods in Roman Dakhleh. Studies in the Indigenous Cults
of an Egyptian Oasis. Groningen: no publisher indicated
Kauffmann H. 1976. Probleme griechischer Säulen. Opladen: Westdeutscher Verlag.
Kemp. B., and P.Rose, 1991. “Proportionality in Mind and Space in Ancient Egypt.”
Cambridge Journal of Archaeology 1: 103–29.
Kienast. H. J. 1985. “Der sog. Tempel D im Heraion von Samos.” Mitteilungen des
Deutschen Archäologischen Instituts, Athenische Abteilung 100: 105–27.
———. 1986/87 “Der Tunnel des Eupalinos auf Samos.” Mannheimer Forum 179–
241.
———. 1991. “Fundamentieren in schwierigem Gelände; Fallenstudien aus dem
Heraion von Samos.” Pp. 123–7 in Bautechnik der Antike, Band 5, ed. A. Hoffmann.
Mainz am Rhein: Philipp von Zabern.
Kienitz, F. K., 1953. Die politische Geschichte Ägyptens vom J. bis zum 4. Jahrhundert vor
der Zeitwende. Berlin: Akademie-Verlag.
Kingsley, P. 1992. “Ezekiel by the Grand Canal: Between Jewish and Babylonian
Tradition.” Journal of the Royal Asiatic Society Third Series, Vol. 2, 339–46.
Kirk, G. S. 1960. “The Structure and Aim of the Theogony.” Pp. 61–95 in Hésiode et
son influence, Entretiens sur l’antiquité classique, Tome 7, ed. K. Von Fritz. Genève:
Fondation Hardt.
———. 1970. “Some Problems in Anaximander.” Pp. 322–49 in Studies in Presocratic
Philosophy, Vol. I, The Beginnings of Philosophy, eds D. J.Furley and R. E. Allen. New
York and London: Humanities Press.
———. Greek Myths, Harmondsworth 1974
Kirk, G. S., J. E. Raven, and M. Schofield, 1995. The Presocratic Philosophers. Cam-
bridge: Cambridge University Press. Reprint of 1957.
Krafft, F. 1971a. Geschichte der Naturwissenschaft I. Die Begründung einer Geschichte der
Wissenschaft von der Natur durch die Griechen. Freiburg: Verlag Rombach.
———. 1971b. “Anaximandros.” Pp. 284–305 in Die Grossen der Weltgeschichte, I, ed.
K. Fassmann, Zürich,
———. 1971c “Anaximander und Hesiodos. Die Ursprung rationaler griechischer
Naturbetrachtung.” Sudhofs Archiv 55:152–79.
264 BIBLIOGRAPHY
Kratzert, T. 1998. Die Entdeckung des Raumes. Vom hesiodischen “cºoV” zur platonischen
“cÔra”. Amsterdam and Philadelphia: B. R. Grüner.
Krischen, F. 1927. Zeitschrift für Archäologischen Bauforschung 77: 23–67.
———. 1956. Weltwunder der Baukunst in Babylonien und Ionien. Berlin: Gebr. Mann
Verlag.
Lambert, W. G., 1975. ‘The Cosmology of Sumer and Babylon.’ Pp. 42–65 in Ancient
Cosmologies, eds. C. Blacker and M. Loewe. London: George Allen and Unwin.
Lamberton, R. 1988. Hesiod. New Haven, Conn.: Yale University Press.
Lamy, L. 1981. Egyptian Mysteries. New Light on Ancient Knowledge. London: Thames
and Hudson.
Lanzone, R. V. 1974. Dizionario di mitologia Egizia. Amsterdam: Benjamins. Reprint
of 1881–86. Torino: Stamperia reale della ditta G. B. Paravia.
Lawrence, A.W., and R. A Tomlinson, 1957. Greek Architecture. New York: Penguin
Books Ltd. Reprinted in1996.
Leclant, J., 1965. Recherches sur les monuments thébans de la XXVe dynastie dite
éthiopienne. 2 vols. Bibliothègue d'Étude, no. 36. Le Caire: Institut français
d'archéologie orientale.
Leclant, J., ed. 1980. Le monde Égyptien. Les Pharaons. Tome 3, L’Égypte du crépuscule.
Paris: Gallimard.
Leibniz, G. W. 1965. Die Philosophische Schriften, Band.7. Edited by C. J. Gerhardt.
Hildesheim: Olms.
———. 1969. Philosophical Papers and Letters. Edited and translated by L. E Loemker.
Dordrecht: Reidel.
Lesher, J. H. 1992. Xenophanes of Colophon: Fragments. Toronto: University of Toronto
Press.
Lévêque, P., and P.Vidal-Naquet, 1997. Cleisthenes the Athenian: An Essay on the Rep-
resentation of Space and Time in Greek Political Thought from the End of the Sixth
Century to the Death of Plato. Translated and edited by D. Ames Curtis. Atlantic
Highlands, N. J.: Humanities Press International.
Lévi-Strauss, C. 1958. Anthropologie structurale. Paris: Plon.
Lippincott, K., ed., 1999. The Story of Time. London: Merrell Holberton.
Livingstone, A. 1986. Mystical and Mythological Explanatory Works of Assyrian and
Babylonian Scholars. Oxford: Clarendon Press.
———, ed., 1989. Court Poetry and Literary Miscellanea. Helsinki: Helsinki University
Press.
Lloyd, A. B. 1975–1988. Herodotus Book II. 3 vols. Leiden: Brill.
Lloyd, G. E. R. 1966. Polarity and Analogy. Two Types of Argumentation in Early Greek
Thought. Cambridge: Cambridge University Press.
———. 1970. Early Greek Science: Thales to Aristotle. New York: Norton.
Bibliography 265
———. 1979. Magic, Reason, and Experience. Studies in the Origin and Development of
Greek Science. Thales to Aristotle. Cambridge: Cambridge University Press.
———. 1983. Science, Folklore and Ideology. Studies in the Life Sciences in Ancient Greece.
Cambridge: Cambridge University Press.
———. 1991. Methods and Problems in Greek Science, Cambridge: Cambridge Univer-
sity Press.
———. 1996. Adversaries and Authorities. Cambridge: Cambridge University Press.
Loraux, N. 1991. “Origins of Mankind in Greek Myths: Born To Die.” Pp. 390–95
in Dictionary of Mythologies (2 vols.), eds.V. Bonnefoy and W. Doniger. Chicago:
The University of Chicago Press.
Mansfeld, J. 1987. Die Vorsokratiker. Stuttgart: Philipp Reclam Jun.
Marcovich, M. 1959. “Was Xenophanes in Paros (Greece), Paros (Dalmatia), or Pharos
(Egypt)?” Classical Philology 54: 121.
———. 1999. “Sources.” Pp. 22–44 in Early Greek Philosophy. ed. A. Long. Cam-
bridge: Cambridge University Press.
Mark, I. S. 1995. “The Lure of Philosophy: Craft and Higher Learning in Ancient
Greece.” Pp. 25–37 in Polykleitos, the Doryphoros, and Tradition, ed. W. G. Moon,
Madison, Wisc.: University of Wisconsin Press.
Martin, R. 1965. Manuel D’Architecture Grecque. Paris: A. et J. Picard.
Mayall, R. N., and M. W. Mayall, 1938. Sundials. Cambridge, Mass.: Sky Publishing
Corporation.
McGlew, J. F. 1993. Tyranny and Political Culture in Ancient Greece. Ithaca and Lon-
don: Cornell University Press.
McKirahan, R. 2001. “Anaximander’s Infinite Worlds.” Pp. 49–65 in Essays in Ancient
Greek Philosophy VI: Before Plato, ed. A. Preus. Albany: State University of New York
Press.
Meißner, B. 1925. “Babylonische und griechische Landkarten.” Klio 19: 97–100.
Menander. 1972. Menandri Reliquiae Selectae. Edited by F. H. Sandbach. Oxford:
Clarendon Press.
Merkelbach, R., and M. L. West, 1967. Fragmenta Hesiodea. Oxford: Clarendon Press.
Mertens, D. 1993. Der alte Heratempel in Paestum: und die archaische Baukunst in
Unteritalien. Mainz am Rhein: Von Zabern.
Mieli, A. 1916. La scienza greca. I Prearistotelici. Vol.I: La scuola ionica, pythagorica ed
eleata. Firenze: Libreria della Voce.
Mitton, S. 1977. Cambridge Encyclopaedia of Astronomy. London: Cape.
Morris, S. 1997. “Homer and the Near East.” Pp. 614–34 in A New Companion to
Homer, eds. I. Morris and B. Powell. Leiden: Brill.
Morrison, J. S. 1959. “The Shape of the Earth in Plato’s Phaedo.” Phronesis 4: 101–19.
266 BIBLIOGRAPHY
Mugler, C. 1953. Deux thèmes de la cosmologie grècque: devenir cyclique et pluralité des
mondes. Paris: Klincksieck.
Mulder, E. 1979. Zon, maan en sterren. Astronomie voor iedereen. Rotterdam: Christofoor.
Murray, O. 1988. “The Ionian Revolt,” in The Cambridge Ancient History (vol. 4).
Cambridge: Cambridge University Press.
———. 1993. Early Greece. London: Fontana Press.
Myres, J. L. 1953. Herodotus, Father of History. Oxford: Clarendon Press.
Naddaf, G. 1998a. “The Atlantis Myth: An Introduction to Plato’s Later Philosophy
of History.” Phoenix 48(3): 189–209.
———. “Lefkowitz and the Afrocentric Question.” Philosophy of the Social Sciences.
28(3): 451–70.
———. 1998c.“On the Origin of Anaximander’s Cosmological Model.” Journal of the
History of Ideas 59(1): 1–28.
———. 2000. “Literacy and Poetic Performance in Plato’s Laws.” Ancient Philosophy
21(4): 339–350.
———. 2001.“Anaximander’s Measurements Revisited.” Pp. 5–21 in Essays in Ancient
Greek Philosophy VI: Before Plato, ed. A. Preus. Albany: State University of New York
Press.
Naddaf, G., and L.Brisson, 1998. Plato the Myth Maker. Translated, edited, and with
an introduction by G. Naddaf. Chicago: University of Chicago Press.
Nagy, G. 1982. “Hesiod.” Pp. 43–73 in Ancient Writers, Vol. I., ed. T. J. Luce. New
York: Charles Scribner’s Sons.
Naydler, J. 1996. Temple of the Cosmos. The Ancient Egyptian Experience of the Sacred.
Rochester, Vt.: Inner Traditions.
Needham, J. 1959. Science and Civilisation in China, Vol.3. Cambridge: Cambridge
University Press.
Neugebauer, O. 1952. The Exact Sciences in Antiquity. Princeton: Princeton University
Press.
———. 1975. A History of Ancient Mathematical Astronomy. 3 vols. Berlin: Springer.
Neuhäuser, I. 1883. Dissertatio de Anaximandri Milesius sive vetustissima quaedam re-
rum universitatis conceptio restituta. Bonnae: Max Cohen et Filius.
Nilsson, M. P. 1932. The Mycenean Origin of Greek Mythology.Cambridge: Cambridge
University Press.
Niwiński, A. 1989. Studies on the Illustrated Theban Funerary Papyri of the 11th and 10th
Centuries B. C. Freiburg, Schweiz: Universitätsverlag Freiburg.
North, J. 1994. The Fontana History of Astronomy and Cosmology. London:Fontana
Press.
Nylander, C. 1970. Ionians in Pasargadae: Studies in Old Persian Architecture. Uppsala:
Acta Universitatis Upsaliensis.
Bibliography 267
Oberbeck, J. A. 1868. Die antike Schriftquellen zur Geschichte der bildenden Künste bei
den Griechen. Leipzig: no publisher indicated.
O’Brien, D. 1967. “Anaximander’s Measurements.” The Classical Quarterly 17: 423–32.
Omlin, J. A. 1973. Der Papyrus 55001 und seine satyrisch-erotischen Zeichnungen und
Inschriften. Torino: Edizione d’Arte Fratelli Pozzo.
Orlandos, A. 1966/68. Les Matériaux de Construction et la Technique Architecturale des
Anciens Grecs. 2 vols. Translated by V. Hadjimichali. Paris: E. de Boccard.
Osborne, R. 1996. Greece in the Making 1200–479 B. C. London and New York:
Routledge.
Owusu, H. 1998. Symbole Ägyptens. Darmstadt: Schirner Verlag.
Pancoucke, C. L. F., ed. 1820 ff. Description de l’Égypte ou recueil des observations et des
recherches qui ont été faites en Égypte pendant l’expédition de l’armée française. Seconde
édition, dédiée au Roi. Antiquités (Planches). 11Vols. Paris: Pancoucke.
———. ed. 1821 ff. Description de l’Égypte ou recueil des observations et des recherches qui
ont été faites en Égypte pendant l’expédition de l’armée française. Seconde édition, dédiée
au Roi. Antiquités-Descriptions. 23 Vols. Paris: Pancoucke.
Parker, R. A. 1974. “Ancient Egyptian Astronomy.” Pp. 51–66 in The Place of As-
tronomy in the Ancient World, ed. F. R. Hodson. Oxford: Oxford Uniocversity Press.
Parker V. 1997. Untersuchungen zum Lelantischen Krieg und verwandten Problemen der
frühgriechischen Geschichte. Stuttgart: Steiner.
Parkinson, R. 1999. Cracking Codes. The Rosetta Stone and Decipherment. Berkeley and
Los Angeles: University of California Press.
Parpola, S. 1993. “The Assyrian Tree of Life: Tracing the Origins of Jewish Monothe-
ism and Greek Philosophy.” Journal of Near Eastern Studies 52: 161–208.
Pedersen, O., and M. Pihl, 1974. Early Physics and Astronomy. London: Macdonald
and James; New York: American Elsevier.
Petrie, W. M. Flinders, 1885. The Pyramids and Temples of Gizeh. New York: Field and
Tuer.
———. 1926. Glass Stamps and Weights: Ancient Weight and Measures. Warminster:
Aris and Phillips.
Petronotis, A. 1972. Zum Problem der Bauzeichnungen bei den Griechen. Athen: Dodona.
Piankoff, A., and N. Rabova, 1957. Mythological Papyri, in Two Parts. New York:
Bollingen Foundation.
Piggott, S. 1992. Wagon, Chariot and Carriage. Symbol and Status in the History of
Transport. London: Thames and Hudson.
Pindar, 1997. Works . Edited and translated by W. H. Race. 2 vols. Cambridge, Mass.:
Harvard University Press.
268 BIBLIOGRAPHY
Pliny the Elder, 1968. Natural History. Edited and translated by H. Rackham. Cam-
bridge, Mass.: Harvard University Press.
Pollitt, J. J. 1990. The Art of Ancient Greece: Sources and Documents. Cambridge: Cam-
bridge University Press.
———. 1995. “The Canon of Polykleitos and Other Canons.” Pp. 19–25 in Polykleitos,
the Doryphoros, and Tradition, ed. W. G. Moon. Madison, Wisc.: Univ. of Wiscon-
sin Press.
Porter, B., and R. L. B. Moss, 1972. Topographical Bibliography of Ancient Egyptian
Hieroglyphic Texts, Reliefs, and Paintings. II. Theban Temples. Oxford: Griffith Insti-
tute.
Powell, B. 1997. “Homer and Writing.” Pp. 3–32 in A New Companion to Homer, eds.
I. Morris and B. Powell. Leiden: Brill.
Prinz, H. 1915. Altorientalische Symbolik. Berlin: Verlag von Karl Curtius.
Raven, J. E. 1951. “Polyclitus and Pythagoreanism.” Classical Quarterly 45: 147–52.
Redshift 1994.(CD-ROM). London: Maris Multimedia.
Rescher, N. 1958. “Cosmic Evolution in Anaximander?” Studium Generale 11: 718–31.
———. 1969. Essays in Philosophical Analysis. Pittsburgh: University of Pittsburg Press.
Reuther, O. 1957. Der Heratempel von Samos. Berlin: Gebr. Mann.
Robb, K. 1994. Literacy and Paideia in Ancient Greece. New York: Oxford University
Press.
Robins, G. 1994. Proportion and Style in Ancient Egyptian Art. London: Thames and
Hudson.
Robinson, J. M. 1968. An Introduction to Early Greek Philosophy. Boston: Houghton
Mifflin Company.
———. 1971. “Anaximander and the Problem of the Earth’s Immobility.” Pp. 111–
8 in Essays in Ancient Greek Philosophy, in eds. J. P. Anton and G. L. Kustas, Albany:
State University of New York Press.
Rochberg-Halton, F. 1983. “Stellar Distances in Early Babylonian Astronomy: A New
Perspective on the Hilprecht Text (HS 229).” Journal of Near Eastern Studies 42:
209–17.
Roller, D. W. 1989. “Columns in Stone: Anaximander’s Conception of the World.”
L’Antiquité Classique 58: 185–9.
Ronan, C. 1973. Lost Discoveries. The Forgotten Sciences of the Ancient World. New York:
McGraw–Hill.
Rosen, R. 1997. “Homer and Hesiod.” Pp. 463–88 in A New Companion to Homer, eds.
I. Morris and B. Powell. Leiden: Brill.
Roth, G. D. 1996. Sterne und Planeten erkennen und beobachten. München: BLV.
Rottländer, R. C. A. 1990. “Zweierlei Maßeinheiten an einem Bauwerk?” Jahreshefte
des Österreichischen archaeologischen Instituts in Wien 60: 19–41.
Bibliography 269
Treue, W., ed. 1986. Achse, Rad und Wagen. Fünftausend Jahre Kultur- und
Technikgeschichte. Göttingen: Vandenhoeck und Ruprecht.
Tritle, L. A. 1996. “Vision or Dream of Greek Origin.” Pp. 303–30 in Black Athena
Revisited, eds. M. Lefkowitz and G. McLean Rogers. Chapel Hill: The University
of North Carolina Press.
Tuchelt, K., ed. 1996. Didyma III: ein Kultbezirk an der Heiligen Straße von Milet nach
Didyma. Mit Beitragen von K. Tuchelt, P. Schneider, T. G. Shattner, und H. R.
Bladus. Mainz am Rhein: Philipp von Zabern.
Usener, H. 1903. “Dreiheit.” Rheinisches Museum für Philologie 58: 1–47, 161–208, and
321–62.
Van der Waerden, B. L. 1974. Science Awakening II: The Birth of Astronomy. New
York: Oxford University Press.
Van Wees, H. 1997. “Homeric Warfare.” Pp. 668–93 in A New Companion to Homer,
eds. I. Morris and B. Powell. Leiden: Brill.
Vernant, J.-P. 1983. Myth and Thought Among the Greeks. London: Routledge and
Kegan Paul.
Vitruvius, 1968. The Ten Books on Architecture. Translated by M. H. Morgan. New
York: Dover Publications Inc.
Vlastos, G. 1970. “Equality and Justice in Early Greek Cosmologies.” Pp. 56–91 in
Studies in Presocratic Philosophy, Vol. I, The Beginnings of Philosophy, eds. D. J. Furley
and R. E. Allen, New York and London: Humanities Press.
Von den Steinen, K. 1897. Unter den Naturvölkern Zentral-Brasiliens. Berlin: D. Reimer.
Von Fritz, K. 1971. Grundprobleme der Geschichte der antiken Wissenschaft. Berlin and
New York: Walter de Gruyter.
Von Gerkan, A. 1959. Antiker Architektur und Topographie. Stuttgart: W. Kohlhammer
Verlag.
Von Steuben, H. 1973. Der Kanon des Polyklet. Tübingen: Wasmuth.
Walcot, P. 1966. Hesiod and the Near East. Cardiff: University of Wales Press.
Waugh, A. E. 1973. Sundials. Their Theory and Construction. New York: Dover Publica-
tions Inc.
Welsby, D. A., 1996. The Kingdom of Kush. London: Thames & Co.
Wesenberg, B. 1976. Review of Petronotis ”Zum Problem der Bauzeichnungen bei
den Griechen.” Gnomon 48: 797–802.
———. 1983. “Beiträge zur Rekonstruktion Griechischer Architektur nach literarischen
Quellen.” Mitteilungen des Deutschen Archäologischen Instituts, Athenische Abteilung 9
(Beiheft). Berlin: Gebr. Mann Verlag.
———. 1985. “Zu den Schriften der Griechischen Architekten.” Pp. 39–48 in
Bauplanung und Bautheorie der Antike, ed. W. Höpfner. Berlin: Deutsches
archäologisches Institut.
272 BIBLIOGRAPHY
(for ancient authors, see also the Index of anachronism, anachronistic fallacy, 172–74,
Classical Passages Cited) 176, 179, 182, 187, 205, 210, 212, 218,
221, 223–25, 237, 242n22, 243n49,
Achilles’ shield, 63n87, 86, 140–41, 173 244n67, 246n97, 246n108, 251n191
Achilles Tatius, 176, 242n27 analogy, 78–79, 89, 92, 153n30, 156n92
Acusilaus of Argos (historian), 57n13 anathyrōsis (technique for fitting column-
adaptation (of animals to the environment), drums), 77, 90–91, 141
15 Anaximander, see also: apeiron, bark, celestial
Adkins, A. W. H., 58n33 bodies, cosmic—, cosmos—, depth in the
aerial view, see map universe, discovery of space, earth,
Aeschylus, 43–44 gnomon, map—, model of the cosmos,
Aetius, 58n31, 251n191 nine, origin—, presteros aulos, sun, three,
Afrocentrism, 67n140 and wheels
Agatharcus (artist), 161n201 book , 9–10, 42, 49–51, 149
Agathemerus (geographer), 32, 35, 62n79 introduction of geometry in Greece, 73
Agenor (son of Libya), 45 rational approach of, 14, 17, 40–41,
agora (public gathering place), 20 49–50
akroterion (roof–ornament), 137–41 symmetry in, 35
Aigyptus (or Aegyptus), son of Belos, on thunder and lightning, 9, 34
brother of Danaus, 44, 66n129 use of analogies, 79
Aigyptus (inhabitant of Ithaca), 64n99 visit to Egypt, 10, 40, 73
Alcaeus (poet), 65n113 visit to Sparta, 33, 140, 152n7
Al Mina in Syria, 21, 26 wrote a Tour of the Earth, 40, 42
alphabet, Anaximander (crater on the moon), 167
and the advent of speculative thought, Anaximander (the historian), identified with
9 Anaximander, the philosopher, 42
Anaximander on the origin of, 51 Anaximander (Milesian author), identified
Egyptian, 46–47 with Anaximander, the philosopher, 43
first datable evidence of, 22 Anaximenes, 65n114, 65n116, 81, 202, 208–
introduction in Greece, 21, 43, 45–46 09, 239
invention of, 21, 47 Andrae, W., 241n3
transmitted by Danaus, 43, 45–48, 50 andres, 12–13
West Semitic, 46 Andrewes, A., 27
Alyattes (king of Lydia), 30, 60n62, 60n64, anthropoi, 12, 15
61n67 anthropomorphism (in Homer and Hesiod),
Amasis (pharaoh), 31, 37, 40, 64n101–02 39, 42
Amphidamas (king of Chalcis), 60n55 apeiron (the Boundless), 167, 237
Amphitres (tyrant of Miletus), 26 as the open space between Earth and
Amun (Egytian god), 75 celestial wheels, 131
273
274 INDEX OF CONCEPTS AND PROPER NAMES
lie behind each other, 168, 181–82, column height, see temple building in Ionia,
201, 208–11, 214–15. 238–40, see proportional rule
also depth in the universe compass, 83
make full circles, 168, 201–02, 207–08, Conant, C., 60n56
238 Conche, M., 37, 57n18, 57n20, 57n26,
order, 181–82, 209–11, 232, 234–36, 62n80, 63n86, 63n92, 68n163, 68n166–
238, 243n49 69, 237, 245n91, 249n159, 249n163
size, 145–46 Copernicus, 214, 240
celestial vault, 169, 180, 208, 210, 215, Corinth, 22, 24
229–31, 236–37, 239–40, see also Cornford, F. M., 237–38, 249n161,
firmament 253n223, 253n225
Cella Building, see Kernbau cosmic architecture, 73–74, 78–82, 86, 89,
Censorinius, 15–16, 58n31 92, 109, 118, 121, 140, 148–50
Chalcis on Euboea, 22, 26 cosmic cylinders (virtual), 170, 173–74, 186,
Charondas in Catana, 21, 59n45 194, 221–28, 241n12, 250n185, 251n195–
Cheiromacha (Labor, party in Miletus), 29 97
Cherniss, H., 50–51, 56n1 cosmic formula, 133–34, 144–49
Chersiphon of Knossos (architect), 80, 125– cosmic geometry, 81
26 cosmic map, see map of the cosmos
Cicero, 33 cosmic module, 118, 144, 148
Cilician Mountains, 68n174 cosmic numbers, 73–151 passim, 171, 211–
Cimmerians, 27–28 18
Clay, D., 163n231 as drawing instructions, 83, 212–14,
Clayton, P. A., 252n198 218, 238
Cleisthenes, 25 indicating distances, 81, 154n48, 212,
Cleomenes (king of Sparta), 33 217
Clère, J. J., 199, 245n95, 246n96 meaning far, farther, farthest, 85,
Codrus (king of Athens), 25, 48 155n62, 215
Coldstream, J. N., 59n43 cosmic proportions, 73–151 passim,
cola-can universe, 252n197 cosmogonical myth, 17
Cole, I. H., 158n128, 161n189 cosmogony as part of Anaximander’s book,
Colophon, 27 10
colonization, 22–23 cosmology (as the intro of Anaximander’s
Coldstream, J. N., 67n141 book), 49
Cole, I. H., 102 cosmos as organism, 89, 100
Coleman, J. E., 67n143 cosmos as temple, 78–80
Colophon, 39, 47 Coulton, J. J., 105, 126, 153n31, 153n36–37,
colonization, 21–22, 27, 31, 73, 75 156n88, 158n140, 159n150–52, 160n167,
column 160n180, 162n226
diameter, see temple building in Ionia, Couprie, D. L., 63n91, 67n158, 68n162,
proportional rule 68n165, 68n172, 152n1, 154n48, 155n62,
height, see temple building in Ionia, 242n28, 244n65, 245n83, 249n165,
proportional rule 249n172, 249n174, 250n178, 250n180,
proportions, see temple building in 250n182, 251n190
Ionia, proportional rule Cratos of Mallos, 163n239
shape of the earth on a Laconian cup, Croesus (king of Lydia), 30–31, 33, 39,
33, 63n86 60n62, 61n71, 63n85, 64n101, 110, 120,
symbolizes seperation of Heaven and 160n168
Earth, 109, 135 cylinders, see cosmic cylinders
column drum Cylon of Athens, 25
in Egyptian temples, 74 Cypselus (king of Corinth), 24, 26–27,
height, see temple building in Ionia 59n51–52, 60n61
proportional rule, 78, 80, 86 Cyrus (king of Persia), 30, 39, 62n73, 102,
as shape of the earth, see earth 160n168
276 INDEX OF CONCEPTS AND PROPER NAMES
heavens, see firmament, and celestial vault symbolic numbers in, 85–86, 155n63–
Hecataeus of Abdera, 44 65, 215, 249n170
Hecataeus of Miletus, 32, 33, 38–43, 45–48, Homeric gods, 16–17
53–54, 62n81, 65n114–15, 83, 245n90 hoplites
Hecate, 87 and the demand for equality, 24
Heidegger, M., 156n93, 247n121 rise of the phalanx, 22–24
Heidel, W. A., 10, 39–40, 42, 50–53, 56n1, Milesian army of, 28
62n80, 63n94, 65n108, 65n112, 65n114– hubris, 19–20
18, 66n122, 67n158, 68n160, 68n168, Hultsch, F., 100–01, 157n113–14, 157n116,
68n170, 69n174–75, 154n52, 174, 194– 157n121, 158n130, 158n135
96, 205, 210, 241n18, 245n82–88, Hurwit, J. M., 37, 39, 61n69, 63n86,
245n90, 246n103, 246n110, 248n136, 65n105, 68n162, 68n164, 68n169
253n220 Huxley, G. L., 26–28, 30, 60n62, 61n69,
Hellen (ancestor of all Greeks), 13, 66n132 62n74, 67n151–52
Hellner, N., 152n1, 156n97–99, 158n133, Hyksos, 43
158n145, 159n159, 161n189
Hendrich, C., 152n1 imagination, 238
Hephaistus, 10, 56n8 infinite worlds, 237–38
Hera, 44, 58n36, 99; see also temple in illo tempore (mythical time), 17, 40
Heracleides of Pontos, 61n69 Io, 44–46, 66n130
Heraclitus, 65n114, 122 Ionic canon, see temple building in Ionia
Heraion, see Temple of Hera Isis. 45–46
hermeneutic circle, 104 isonomia (equal rights), 27, 31
Hermes, 44, 56n8 Israël, G., 246n100
Herodotus, 40, 43. 45–48, 245n90 Ister (= Danube), 63n94, 68–69n174
Herrmann, A., 245n87 Iverson, E., 161n204, 162n208
Heroology (as one part of Anaximander’s
book), 40, 42, 46, 48 Jacob, C., 62n82
Hesiod, 33, 40, 42, 51, 58n39, Jacoby, F., 154n52
and the advent of writing, 21 Jaeger, W., 154n40
aim of the Theogony, 10 James, 153n18
cosmology, 208 Jeffrey, L. H., 22, 25–26, 59n44, 60n56,
on the distance of the heavens, 215 60n60, 61n66–67, 61n69, 61n71, 62n75,
on justice, 19–20 62n77
misogyny, 56n8 Johnson, F. R., 253n228
on the Pleiades, 180–81, 243n43 Jucker, H., 63n86
symbolic numbers in, 85–86, 155n67– Junker, H., 252n203
68, 211, 215, 249n170 justice, see dike
Theogony as cosmogonical myth, 16, 18
Works and Days advocating social Kahn, C. H., 57n28, 58n32, 154n40, 154n44,
reform, 18 154n46, 154n55, 156n108, 163n235, 175,
Heubeck, A., 64n99 205, 214, 237, 242n26, 244n64, 244n67,
hexameter, 122–23 245n93–94, 246n113, 247n132, 248n141,
Hieronymus of Rhodes, 152n1 249n161–62, 249n164, 250n177, 250n185,
Histiaeus (tyrant of Miletus), 29, 62n74–75 251n191, 253n221
Hipparchus of Nicaea, 51 Kalchas (prophet), 85
historia, Anaximander’s book as, 49–50 Kambartel, F., 246n116
Hölscher, U., 250n176 Kaper, O. E., 252n207
Holloway, R. R., 161n197, 161n202 Kauffmann, H., 183–84, 243n54
Homer, 32, 35, 42, 62n81 Kemp, B., 154n49
conception of the heavens, 170, 202, Kernbau, see temple building in Ionia
208 Kienast, H. J., 98, 152n1, 153n23, 153n28,
on the shape of the earth, 34 154n54, 156n100–01, 158n141, 162n215
Index of Concepts and Proper Names 279
Naddaf, G., 25, 31, 56n2, 56n4, 57n12, Oberbeck, J.A., 161n197
59n53, 65n107, 67n140, 73, 152n1, O’Brien, D., 212, 214–15, 225, 248n152,
152n5, 154n46, 176, 196, 242n30, 248n155–56, 249n160, 251n194
245n90 Ocasia, C. J., 152n1
Nagy, G., 18 Ocean (or Oceanus), 34–35, 52, 63n87, 91,
Napata (town in Nubia), 75 194
NASA, 246n100 Odysseus, 23, 85
Naucratis (Greek trading post in Egypt), 10, Oedipus, 12
36–37, 39–40, 64n104, 65n113, 67n145, oikoumene, see map of the world
73, 75, 77 Olympia, 137–39
navel of the world, 34–35, 49, 55, see also Omlin, J. A., 233, 252n206, 252n208
Delphi, Delos, Egypt, Miletus, and orbis terrarum, 50
Babylon organic analogy, 89, 98–100
Naxos, 29 origin of the cosmos, 140
Naydler, J., 241n2, 253n209 origin and development of humanity, 9–10,
Neco (I), (or Necho, pharaoh), 28, 63n90, 15–17, 40, 49, 57n32
63n95, 64n102, 75 as born from the earth, 10–11
Needham, J., 244n74 as fashioned by a divine artist, 10
Neferkaure, see Shabaka mythical accounts, 9–13
Nefertem-khu-Re, see Taharqa rational approach, 14, 17, 40–41, 50
Neleids (group inhabitants of Miletus), 26– origin and development of society
27 (politogony), 10, 32–43, 49
Neleus (son of Codrus, founder of Miletus), mythical accounts of, 9, 17–20
25–26, 48 origin of human reproduction, 11–13
Neugebauer, O., 68n159, 247n133 origin of land animals, 13
Neuhäuser, I., 220, 250n182 origin of language, 51
Index of Concepts and Proper Names 281
Vidal-Naquet, P., 63n88, 63n91, 68n165 Western world-picture, see Western paradigm
Vitruvius, 105, 108–09, 125, 139, 153n34, of the universe
159n165 wheels, 167, 170–71, 174–78, 201–02, 208,
Vlastos, G., 155n69 211, 215, 218–28, 236, 239, 242n30,
Van Soldt, W. H., 247n133, 254n231 242n31, 249n167, 250n185, 251n195, see
Von Fritz, K., 247n123–24 also celestial bodies, and cosmic numbers
Von Gerkan, A., 130, 131, 162n214 Wildung, 152n12
Von den Steinen, K., 88, 155n83 Willets, R. F., 58n42
Von Steuben, H., 128 women (creation of ), 12, 56n8
vortex, 141, 241n18 writing
advent in Greece, 21, 46
Walcot, P., 56n8 and the advent of speculative thought,
Waugh, A. E., 244n60 9, 31
Wealth (party in Miletus), 29 introduction in Greece, 45–46
Welsby, 152n12 Phoenician system, 46
Wesenberg, B., 86, 105–06, 105–10, 153n37,
154n54, 155n73, 159n148, 159n153, xenia (guest-friendship pact), 61n67
159n158–60, 159n164–66, 160n167 Xenophanes, 38–40, 42, 45, 65n112
West, M. L., 30–31, 45, 62n76, 62n78,
66n135, 211, 215, 218, 225, 247n131, Yalouris, N., 63n86, 121, 137–40, 162n228,
247n135, 248n142–43, 249n169, 162n230, 163n232, 163n234
251n191–92
West, S., 64n99 Zaleukos (or Zaleucus) in Locri, 21, 59n45
Western concept of space, see Western Zeus, 18–19, 35, 44–45, 58n36, 58n39
paradigm of the universe zodiac, 174, see also ecliptic
Western paradigm of the universe, 167–68, Zodiac of Dendara, 234, 236
172, 201, 238–40
Index of Concepts and Proper Names 285
(references to pages within a work are DK 12A1 (1), 243n48, 243n51, 243n56
italic, references in Anaximander in DK 12A1 (2), 242n33, 242n34,
Context are roman) 245n92, 248n157
DK 12A1ff., 152n2
DK 12A2, 152n8, 242n32, 242n34,
Acusilaus 243n56, 244n70, 245n92,
DK 9B33, 56n11 248n157, 248n158
Aeschylus DK 12A3, 152n6
Eumenides DK 12A4, 63n84, 243n56, 244n70
81–83, 71 DK 12A5, 163n237, 242n19
Prometheus Bound DK 12A5a, 63n83, 152n7
460–61, 67n148 DK 12A6, 62n79, 62n81, 242n33,
Suppliants 244n77, 245n92
270, 64n97 DK 12A7, 242n21
270ff., 66n131 DK 12A8, 253n227
281ff., 66n131 DK 12A9, 253n227
323, 66n131 DK 12A10, 57n22, 155n74, 163n240,
485, 64n97 241n15, 248n137, 248n138
496–98, 66n131 DK 12A11, 31, 57n16, 57n19, 57n23,
584, 67n153 63n89, 163n241, 241n11, 242n23,
2420, 44 246n102, 248n144
Aetius DK 12A11 (5), 243n47, 248n148,
de plac.phil. 249n167
II.20.1, 163n233 DK 12A17, 238, 249n161
II.21.1, 154n41, 154n58 DK 12A18, 241n14, 242n23, 243n45,
II.25.1, 154n42 248n151, 250n177, 250n188,
II.15.6, 154n43 251n195
Agathemerus DK 12A20, 243n43
Geography DK 12A21, 241n14, 242n23, 242n25,
1.1, 62n79 242n27, 243n47, 248n148,
1.1–2, 154n52 248n149, 248n154
1.1.2, 34 DK 12A22, 225, 241n14, 242n23,
Anaxagoras 242n25, 243n48, 248n150,
DK 59A1 (9), 251n193 250n178
DK 59A8, 163n232 DK 12A23, 31, 63n89
DK 59A67, 251n193 DK 12A24, 63n89
Anaximander DK 12A25, 241n11, 249n167
DK 12A1, 52, 66n124, 152n7, DK 12A26, 31, 246n101
163n236, 244n70, 244n77 DK 12A27, 57n16, 57n21
286 INDEX OF CLASSICAL PASSAGES CITED
50
50,39,39
100
PHILOSOPHY
Anaximander in Context
100
Promoting a new, broadly interdisciplinary horizon for future studies in early Greek
100
philosophy, Dirk L. Couprie, Robert Hahn, and Gerard Naddaf establish the cultural con-
100
text in which Anaximander’s thought developed and in which the origins of Greek phi-
RED
achievement, the authors call our attention to the historical, social, political, technological,
BLU
arguments that help to illuminate Anaximander’s thought, in particular the need to com-
Anaximander in Context
prehend the culture of Archaic Miletos, the strong role of architecture as an essential
75
element to understanding Anaximander (not a new view but one here carried further
50
than ever before), and the need not to dismiss Anaximander’s stranger views on astronomy
25
the Architects:The Contributions of Egyptian and Greek Architectural Technologies to the Origins
of Greek Philosophy, also published by SUNY Press. Gerard Naddaf is Associate Professor
50
and Chair of the Department of Philosophy at York University. He is the translator and
25
www.sunypress.edu
50
25
100
100
100
SUNY
100