Ziggurats, Colors, and Planets: Rawlinson Revisited

Full text

57 JCS 60 (2008)
The Walls of Ecbatana
Since at least the mid-nineteenth century, scholars have pondered the meaning of the colors, which,
according to Herodotus (Hist. 1.98), adorned the seven concentric battlements of the royal city of Ecbatana
in Media built by King Deioces in the late-eighth century BC:
The building was so contrived that each circle of walls is higher than the next by the battlements only. The fact that
the place chosen was itself a hill helps the design, but it was also much strengthened by contrivance. The circles of
the walls were, in all, seven, and within the ˜nal circle are the royal palace and the treasuries . . . The battlements
of the ˜rst circle are white, the second black, the third scarlet, the fourth blue, the ˜fth orange. Thus the battle-
ments of those ˜ve circles are painted with colors; but of the last two circles, the one had its battlements coated
with silver, the other with gold. (trans. Grene 1987: 80–81)1
Hence the color sequence was:
white – black – dark red – blue – light red – silver – gold
Sir Henry Rawlinson (1810–1895), traditionally the founding father of Assyriology, seems to have
been the ˜rst to suspect an astronomical signi˜cance in these colors, proposing that each represented
one of the seven traditional planets. His primary source for these associations was the twelfth-century
Azerbaijani poet Nizami:2
This [Herodotus’ account] is manifestly a fable of Sabaean origin, the seven colours mentioned by Herodotus being
precisely those employed by the Orientals to denote the seven great heavenly bodies, or the seven climates in
which they revolve. Thus Nizámí, in his poem of the Heft Peïker, describes a seven-bodied palace, built by Bahrám
Gúr, nearly in the same terms as Herodotus. The palace dedicated to Saturn, he says, was black—that of Jupiter
1. In Greek, the respective colors are: leukós, “white,” mélas, “black,” phoiníkeos, “dark red,” kˆanos, “blue,” sandarákinos, “light
red,” katargyroménos, “covered with silver, silvered” and katakechrysoménos, “gilded.” For phoiníkeos, LSJ s.v. “foinÇkeoÍ”) give
“purple-red, crimson, and (generally) red,” as well as “dark red,” from phoinix with the same meanings (p. 1948). Rawlinson (1858:
242) and Grene (above) legitimately translate it as “scarlet.” As for sandarákinos, Liddell and Scott (1996: 1582 s.v. “sandravkh”)
give “of orange colour.” The adjective is derived from sandaráke, “red sulphide of arsenic, realgar” or “an orange pigment made
therefrom”; cf. Latin sandaraca, “a red coloring matter” (Lewis and Short 1879: 1626). We have opted here for a simple distinction
between phoiníkeos as dark red and sandarákinos as light red.
2. In Nizami’s romance Haft Paikar (“The Seven Beauties”), the hero Bahram meets seven maidens on seven successive
days. Each has a diˆerent color of eyes, garments, and sometimes hair, and occupies a pavilion with a corresponding color. The
colors are linked to the planets governing the days of the week. For translation and commentary, see Wilson (1924).
ZIGGURATS, COLORS, AND PLANETS:
RAWLINSON REVISITED
Peter James and Marinus Anthony van der Sluijs (London)

PETER JAMES AND MARINUS ANTHONY VAN DER SLUIJS58
orange, or more strictly sandal-wood colour (Sandalí)—of Mars, scarlet—of the sun, golden—of Venus, white—of
Mercury, azure—and of the moon, green—a hue which is applied by the orientals to silver. (Rawlinson 1841: 127–28)
He developed his views in 1854 when excavating parts of the great ziggurat temple of Nabû at
Birs Nimr¿d (ancient Borsippa). According to a cylinder he discovered, the ziggurat had been reno-
vated by Nebuchadrezzar II (604–562 BC). The cylinder states that the ziggurat was called
é.ur4.(me).imin.an.ki, “house which gathers the seven (me’s) of heaven and underworld” (George
1993: 157 s.v. #1193). Rawlinson took these mes to be “the planets of the seven spheres” (1861: 17–18).3
In keeping with this, he set out to identify seven stages in the ruins of the structure and satis˜ed him-
self that each had been decorated with a diˆerent color. He reconstructed these, in ascending order, as:
black – rich red brown – bright red – gold – yellow – blue – silver
Then, using the “Sabaean” associations he had derived from Nizami, Rawlinson recognized in this
sequence “the well known [planetary] order of Saturn, Jupiter, Mars, Sol, Venus, Mercury, and the
Moon . . .”:
For Rawlinson this was proof that the Babylonians were not only interested in the order of the planets,
but that they had arrived at the most accurate system possible within the limits of a geocentric perspec-
tive.4 The design of both Birs Nimr¿d and Ecbatana apparently presupposed knowledge of this planetary
sequence. Rawlinson felt that “hints” from a tradition about the variegated ziggurat of Borsippa could
have fed into the story of the walls of Ecbatana (1861: 18 n. 1). However, as the reconstructed sequence
at Borsippa did not match that given by Herodotus for Ecbatana, he was forced to conclude that the re-
ported color order was irrevocably muddled.5
The Borsippa Ziggurat
But what did Rawlinson actually ˜nd at Birs Nimr¿d? Starting with the lowest level of the ziggurat
he reported the following:
1. The well-preserved lowest tier was “thickly coated with bitumen,” providing the black color as-
sociated with Saturn (Rawlinson 1861: 18).
2. Burnt bricks of high quality, “a rich, red brown.” Despite his admission that “it is not very certain
what color we are to attribute to Jupiter,” the logic of his wider assumption led Rawlinson to
associate this stage with that planet (1861: 19).
3. Brighter red bricks, which to Rawlinson indicated Mars (1861: 9, 19–20), universally recognized
in the ancient world as “the red planet.”
3. The meaning of the word me is among the most debated of the Sumerian lexicon. Terms such as “power” or “decree” are
currently preferred; for a discussion see Averbeck (2000: 418 n. 2).
Saturn Jupiter Mars Sun Venus Mercury Moon
black rich red brown bright red gold yellow blue silver
4. The outer planets (Saturn, Jupiter, and Mars) are given in their correct order, starting with the most distant from the
Earth, and are correctly separated from the inner planets Venus and Mercury. The Moon is in its correct position as the nearest
body to the Earth.
5. His brother George experimented with a weekday sequence of planets for Ecbatana but similarly had to conclude that
Herodotus had “accidentally reversed the places of black and white, and of scarlet and orange” (1858: 242).
One Line Short

ZIGGURATS, COLORS, AND PLANETS: RAWLINSON REVISITED 59
4. Pinkish bricks, which Rawlinson thought were originally clad in gold, representing the Sun,
but subsequently plundered; he found the fourth stage severely damaged, as if by pickaxe
(Rawlinson 1861: 20).
5. Yellow bricks. Without specifying his sources (evidently Islamic), Rawlinson (1861: 21) opined
that “Venus was ˜gured in the temple of Borsippa as a light yellow.”
6. Gray. Using a highly elaborate argument (see below), Rawlinson (1861: 22) argued that this
level had originally been encased in a coating of burnt blue slag. He asserted that “The sphere
of Mercury . . . is everywhere represented as blue.”
7. Gray, weather-beaten bricks. Rawlinson (1861: 23–24) explained the absence of a silver coating,
corresponding to the Moon, again by plundering.
Planetary associations aside, Rawlinson’s own words reveal the highly prejudiced nature of his de-
cisions regarding the “original” colors by which the stages were distinguished. Rawlinson admitted
particular problems regarding the great pinnacle of brickwork at the top of the ruin, thought to repre-
sent the sixth and seventh levels, plus a presumed chapel on the summit:
It may be objected that the whole extent of the standing pile exhibits, at present, one uniform appearance of dark,
weather-beaten brickwork, and that there is no trace of its having been divided into two stages, or having sup-
ported a superstructure. (1861: 22, 23)
Rawlinson’s explanation for the absence of the expected color (blue) from the sixth stage (the base
of the “pile”) was even more ingenious than that for stages #4 and #7. Here he appealed to “the large
detached masses of vitri˜ed matter, now cumbering the upper platform” (1861: 22). Described by him
as “blue slag,” more of it was “strewn about the surface of the mound, and in some instances . . . rolled
down into the plain” (1861: 7). In Rawlinson’s opinion, it “most unmistakably split oˆ from the lower
portion of the pile” and hence originally formed the exterior casing of the sixth stage (1861: 22, 7). The
reasoning involved the idea that this stage of the ziggurat had been deliberately ˜red: “all this portion
of the building had been arti˜cially vitri˜ed at the time of its construction,” the intense ˜re converting
its exterior into “one uniform mass” of blue slag which had subsequently split oˆ “under the action of
the elements” (1861: 6, 7).
The weakness of these arguments is self-evident. The idea that the ˜re was due to a deliberate pro-
cess, intended to convert the sixth stage into “slag,” is extraordinary. Such a method is without parallel
in the architectural history of ancient Mesopotamia.6 As observed by Julius Oppert (1863: 202), who
examined the structure shortly before Rawlinson, the intensity of the ˜re had been enough to warp
some layers of brickwork to the extent that they were no longer horizontal. The burning must thus
surely have been unintentional on the part of the builders.
Extensive merging of the ostensible brick colors in the interior clearly caused Rawlinson (1861: 8,
21) considerable di¯culty. For example:
The pink and yellow layers are so intermingled, where the zones, as exposed in the trenches, appear to join, and
generally, indeed, wherever the bricks can be examined around the slope of the mound, that it is impossible to say
exactly where one division ends, or the other begins.
Likewise, Rawlinson’s distinction between the red brickwork of layers 2 and 3 (see above) was hardly
compelling.
6. Rawlinson may have been in˘uenced by an idea widespread in the late-eighteenth century, according to which the so-
called vitri˜ed forts of Scotland had been deliberately ˜red, a proposition laid to rest in Wilson (1851: 413–18).

PETER JAMES AND MARINUS ANTHONY VAN DER SLUIJS60
Indeed, so ruinous was the condition of the ziggurat that it seems extremely doubtful that Rawlinson
was able to examine any exterior surface (as opposed to interior brickwork) as it was intended by its
builders, apart from the bitumen coating on the ˜rst tier (Koldewey 1911: 57; see also Allinger-Csollich
1998: 103) and the blue-glazed bricks that were not found in situ. Because of this, Robert Koldewey
(who carried out the ˜rst properly scienti˜c excavations at Babylon in 1902) concluded from his visit to
Birs Nimr¿d that the alleged color distinction between the stages was nothing but fantasy (1911: 58).
Most recently, the Austrian excavators working at Borsippa between 1980 and 1996 (Allinger-Csollich
1998) have not discerned any signi˜cant diˆerence in the colors of the core brickwork. Aside from the
black and the blue exterior decoration, it seems clear that the color diˆerences perceived by Rawlinson
in the interior were due to the accident of using diˆerent batches of bricks, varying in color either
because of the clay source used and/or the degree of ˜ring.
To conclude, it appears that wishful thinking misled Henry Rawlinson in his interpretation of Birs
Nimr¿d, reading Nizami’s colors into the ruins. The only colors supported by archaeological evidence
are the black and blue of the exterior decoration. That the latter was clearly not produced by ˜ring a
whole stage of the ziggurat follows from Nebuchadrezzar’s cylinder, which, though fragmentary, states
with respect to the ziggurat at Babylon: “with glazed bricks of pure blue color I raised it to its summit”
(trans. Beaulieu 2000: 310).7 The same text, among others, also con˜rms that the Babylonians decorated
temples with precious metals: “I applied shining gold (glaze) instead of plaster. Ezida I built anew
and with silver, gold, choice gems, copper, musukkanu-wood (and) cedar-wood, I completed its work.”
Additional examples could be supplied.
The Rise and Fall of Astral Interpretations
Despite the manifest ˘aws in Henry Rawlinson’s understanding of Birs Nimr¿d, it was accepted
quite uncritically by many leading orientalists of the late-nineteenth to early-twentieth centuries.8 A
paradigm had been set for the interpretation of ziggurats. Henry Rawlinson’s notion that the Babylo-
nians possessed advanced knowledge of the planets was promulgated by his brother George Rawlinson
(1879: 571–79) and spawned a plethora of “astral” interpretations of Mesopotamian myth and re-
ligion. By the early 1900s, these were ˘ourishing in a—largely German—movement referred to as Pan-
Babylonianism (Schmidt 1972: 97–102; Parpola 2004). Its central tenet was that the Babylonians were
extremely accomplished astronomers, who had invented the zodiac as early as 3000 BC and possessed
not only sophisticated mathematical learning and an accurate calendar, but knowledge of such matters
as the phases of Venus and even the precession of the equinoxes.
Notable early critics of Rawlinson included Jensen, who objected that the alleged Borsippa cor-
respondences could not be meaningfully applied to the sequence of decorative colors found on the
Khorsabad ziggurat (1890: 143; see below); and Jastrow, who outrightly rejected any association of these
colors with planets (1898: 618). Scepticism appeared to be warranted when, only a few years after-
wards, Koldewey’s inspection of the ruins of Birs Nimr¿d repudiated Rawlinson’s claim that diˆerent
colors could be discerned in the brickwork of the various stages.
By the mid-twentieth century, mention of Rawlinson’s planetary color scheme for Borsippa was
fading from the literature, and along with it the very idea that there was a link between seven-staged
ziggurats and the planets. More broadly, Pan-Babylonianism turned out to be a remarkably short-lived
trend in scienti˜c history. By the early-twentieth century, it was failing to sustain broadsides from a
number of critics, including, most notably, the brilliant Jesuit scholar Franz Xavier Kugler (1907). Its
7. Whether this passage implies that blue-glazed bricks decorated the actual summit of this building is unclear.
8. E.g., Budge (1884: 13–14); Sayce (1887: 115); Jeremias (1911: 17).
One Line Long

ZIGGURATS, COLORS, AND PLANETS: RAWLINSON REVISITED 61
excesses of interpretation were rightly exposed, but the resulting embarrassment meant that scholar-
ship drew in its horns generally with respect to astral interpretations of Mesopotamian religion. It has
yet to fully recover from the backlash (see Parpola 2004).9
The Ziggurats of Khorsabad and Ur
The issue might have been entirely laid to rest, were it not for the fact that Herodotus’s sequence of
colors appears to have been largely con˜rmed by archaeological work at the sites of Khorsabad and Ur.
Sargon II (721–705 BC) founded Khorsabad (D¿r †arrukÿn) in Assyria.10 A French team that ex-
plored its ziggurat in the mid-nineteenth century uncovered four surviving levels. The excavators, Vic-
tor Place and Felix Thomas, discovered that these stages were still partly covered with colored stucco—
the lowest stage white, the second black, the third reddish purple, the fourth blue (1870: 79). Among
the ruins, they reported numerous fragments of enamelled bricks colored vermilion, silver-gray, and
gold. They took this as proof that there had been three further stages to the ziggurat, decorated with
these colors.11 It did not go unremarked that the whole array matches the very colors listed by Herodotus
(Babelon 1906: 74–75), allowing that vermilion (a vivid red or orange-red) is equivalent to Herodotus’s
sandarákinos (or light red—see n. 1 above).12 The precise order in which the hypothetical upper stages
would have appeared remains conjectural, but it seems reasonable to suppose that they continued He-
rodotus’s sequence, ˜rst because of the good match provided by the colors of the four surviving levels,
and second because the placement of the most expensive decoration—silver and gold on top—makes
good sense in purely practical terms, making it extremely plausible that the vermilion stucco belonged
to the ˜fth tier.
Place and Thomas’s observations were largely con˜rmed by the twentieth-century excavations of
Khorsabad. Henri Frankfort, one of the excavators, gave the colors of the extant stages as follows:
There were actually three stages, and part of the fourth was preserved. . . . Each of them was eighteen feet high
and decorated with recesses; each was painted a diˆerent colour: the lowest white, the next black, the third red,
and the fourth white. (1970: 78–79)
As for the last, Frankfort allowed that this was: “Perhaps bleached blue, for the succession of colours of
the three lowest stages correspond with Herodotus’s report on the tower of Babylon [sic], where the
fourth stage was blue.”13 Frankfort was evidently working from his own observations, for had he relied
on Place and Thomas solely he would have simply given blue for the fourth color.
9. Perhaps the nadir of astral interpretation was Jacobsen’s understanding that the deity Inanna originated as the “numen
of the date storehouse” (1976: 135), with her well-documented astral aspects (relating to the planet Venus, the star Sirius, and
cometary imagery) being of secondary importance. More recent research has shown that the association of Inanna with a celestial
body, probably Venus, is apparent not only from the early-second millennium BC (Heimpel 1982: 10–11) but from the very earliest
Sumerian (Archaic) texts of the late-fourth millennium BC (Szarzynska 1993: 9–10).
10. Interestingly, Deioces of Media, the traditional founder of Ecbatana, may well have been a contemporary of Sargon II.
For the various dates ascribed to Deioces see Scurlock (1990); Henige (2004).
11. “Chacun des sept étages de l’Observatoire était peint d’une couleur particulière. La peinture des quatre premiers étages
se voyait encore sur la ruine; celle des trois autres a été restituée, avec les étages eux-mêmes, et suivant les données que nous
avons déjà fait connaître. Les sept couleurs, blanche, noire, rouge, bleue, vermillon, argent et or, sont de larges teintes plates,
hautes chacune de 6 mètres . . .” (Place and Thomas 1870: 79).
12. Place and Thomas must have had reasons to opt for such a technical word, rather than “orange” or “rouge,” especially
given that Thomas was an artist. Vermilion represents the brighter end of the red spectrum, gravitating towards the orange. The
analogy with Herodotus’s sandarákinos is therefore close.
13. Herodotus, of course, did not report colored stages for the ziggurat of Babylon. Frankfort’s uncharacteristically sloppy
remark here is compounded by another in the same volume, where blue has even changed position (1970: 52): “Babylon, where
Herodotus observed that each of the seven stages had a diˆerent colour, the uppermost being blue.” A similar mistake was made
by Woolley (1954: 219).

PETER JAMES AND MARINUS ANTHONY VAN DER SLUIJS62
Evidence of colored stages has also been retrieved from the ziggurat at Ur. In its original form, as
constructed by Ur-Namma and his son †ulgi (ca. 2100–2050 BC) of the Third Dynasty of Ur, the tower
consisted of four stages. According to its excavator, Sir Leonard Woolley, the Neo-Babylonian king
Nabonidus (555–539 BC) expanded the building into seven stages, leaving only the ˜rst level of the
original intact (1925: 219; 1939: 135–43).
In his initial report, Woolley summed up his ˜ndings:
The shrine, as we have seen, was bright blue, shining in the sun. The top was red: it was built of large lightly ˜red
bright red bricks and was covered with plaster of the same colour. Below this the whole ziggurat, walls and steps
alike, was black, the brickwork covered with a thin coat of bitumen applied with a brush. Below this again was the
white-washed columned wall of the court. (1925: 14)
Here Woolley connected the blue bricks with a shrine forming the fourth stage but in his ˜nal report
(1939: 141) the blue is assigned to the shrine at the hypothetical seventh stage of the ziggurat. The matter
is inconclusive because the bricks were not found in situ. In an accompanying note, Gadd invited com-
parison with the colors found at Khorsabad, con˜rming en passant the validity of Place’s reconstruction
(in Woolley 1925: 14 n. 1 and 1939: 142 n. 1).
Ecbatana: A Fresh Approach
Although Rawlinson’s analysis cannot in any way be used as a yardstick by which to assess the
value of Herodotus’s Ecbatana account, the archaeological evidence from Khorsabad and Ur warrants a
reinvestigation of a possible connection between ziggurat colors and planets—especially as the planet-
color associations argued by Rawlinson have left an indelible mark on the literature, notably in disci-
plines other than Assyriology and despite the decline of Pan-Babylonianism (see, e.g., Golzio 1983: 56–
57). The matter is somewhat tangled, but one that we trust is worth the unravelling. In the following
pages, the respective presuppositions of Rawlinson’s argument will be reexamined in this order:
in architecture:
•whether it is legitimate to compare the legendary “battlements” (promacheonés) of the royal
citadel complex of Ecbatana to the stages of a ziggurat;
•whether some ziggurats consisted of seven stages;
•whether Babylonian architects, at least at some point, worked with a prescribed color system in
the decoration of ziggurats;
in astronomy:
•whether Babylonian astronomers had a concept of seven planets, grouping the Sun and Moon
together with the ˜ve “proper” ones;
•whether they assigned a speci˜c symbolical color to each of the planets;
and hence:
•whether the respective stages of certain Babylonian ziggurats represented planets.
To begin with the ˜rst, the common assumption since Rawlinson’s time has been that the tradition of
the seven-colored battlements of Ecbatana may have arisen from con˘ation with the tiers of a ziggurat
(see, e.g., How and Wells 1912: 104; Singor 1992: 408). From an archaeological perspective, the battle-
ments story has always seemed unlikely. Moreover, there was a long tradition of ziggurat building in
the Iranian world, of which a striking example is that at Choga Zanbil (thirteenth century BC) in Elam
(Potts 1999: 223–26). The tradition continued through Neo-Assyrian times (e.g., the ziggurat at Susa
described and depicted by Assurbanipal; Parrot 1949: 47–48), while the tomb of the sixth-century

ZIGGURATS, COLORS, AND PLANETS: RAWLINSON REVISITED 63
Persian king Cyrus (allegedly a descendant of Deioces) is like a miniature seven-tiered ziggurat, with
the temple-shaped tomb placed on a series of six plinths (Parrot 1949: 50–51). In this light, it is legiti-
mate to compare the seven colors given by Herodotus for the battlements of Ecbatana to the seven-
colored stages of ziggurats.
The Seven-Tiered Ziggurats
Turning to the second question, Gadd (in Woolley 1925: 14 n. 1 and 1939: 142 n. 1) countered Raw-
linson’s planetary theory with the objection that ziggurats often had less than seven stages, commonly
three or four. Indeed they did, but this does not preclude the possibility that ziggurats with seven stages
may at some point have exempli˜ed an idealized prototype. No intact examples of ziggurats have been
found, and, as we have seen, their poor condition typically does not permit con˜dence about the original
number of stages. Nonetheless, in some excavated examples of the ˜rst millennium BC, it has been sur-
mised that seven was the original number of tiers. At Borsippa, for example, Allinger-Csollich counted
˜ve stages in the ruins of Nebuchadrezzar’s restoration, but allowed for the possibility of two additional
top stages that left no archaeological remains (1998: 103). At Ur, Woolley discovered three or four stages
(depending on whether the white court is included in the count), but, on the basis of architectural pro-
jections, he suggested that Nabonidus’s intention had been to transform the existing structure into one
of seven stages.14 At Khorsabad only four stages were discerned, but Frankfort thought it reasonable
that there had been three further stages, as the height of the structure would then have matched the
base “and this was, according to Strabo, the case in Babylon” (1970: 79).15
Reconstructions aside, what do iconography and texts tell us about the number of stages? Parrot’s
survey includes numerous illustrations, for example, a three-tiered structure on an Old Babylonian
cylinder seal, four- to ˜ve-tiered towers on Assyrian cylinder seals, and the ˜ve-tiered ziggurat of Susa
on a relief sculpture depicting Assurbanipal’s Elamite campaign (1949: 37–50). In addition, a consider-
able body of literary and pictorial evidence documents the concept of a ziggurat with seven stages:
1. The so-called Esagila tablet from Uruk (AO 6555), ˜rst noticed by George Smith in 1876, dates to
229 BC (Parrot 1949: 22–24.; Wiseman 1991: 71) and is now also known from a partial duplicate
in the British Museum (BM 40813); they are thought to be copies of an original no later than
the early-seventh century BC (George 2005/2006: 75, 78; Allinger-Csollich 1998: 290–94). This
sets out the dimensions of the temple tower Etemenanki (é.temen.an.ki) of the Esagila com-
plex in Babylon, which was completed by Nebuchadrezzar II. Seven stages are described, with
their respective dimensions (AO 6555 ll. 36–42). The height of the structure is equal to the base,
a fact supporting Strabo’s statement and Frankfort’s deduction regarding the original height of
the ziggurat at Khorsabad. George (2005/2006: 77, 86) regards it as “an ideal of how the tower
was meant to look,” not “a physical survey of a built structure.”
2. A broken Neo-Babylonian stele (MS 2063) in the Schøyen Collection depicts a king, apparently
Nebuchadrezzar II, “standing before a seven-storey ziqqurrat drawn in outline and labelled
[é].tem[en].an.ki [z]i-qú-ra-at babili (ká.dingir.ra)ki ‘E-temen-anki, ziqqurrat of Baby-
lon.’ The tower is as high as it is wide.” (George 2005/2006: 79).
14. Woolley’s considerations (1939: 137–41) included the distinct changes made by Nabonidus’s builders to the lower levels,
calculations from the amount of debris covering the ruins, suggesting that it originally stood to a considerable height, the layout
of the ˘ights of stairs and their angles, plus an idealized plan providing the best symmetry possible.
15. See Strabo, Geog., 16.1.5, where the “tomb of Belus” (Belou táphos) is described as a quadrangular structure a stadion
both in length and height.

PETER JAMES AND MARINUS ANTHONY VAN DER SLUIJS64
3. Around 430 BC, Herodotus (Hist. 1.181.3) reported eight stages for the temple of “Zeus Belus”
in Babylon, which can only have been the Etemenanki. Comparison with #1 and #2 shows
that Herodotus’s count was inaccurate, but it is easy to think of plausible explanations for the
apparent discrepancy (George 2005/2006: 76 n. 2; Drews 1973: 180 n. 118).
4. A Late Babylonian tablet from Babylon (BM 38217; Wiseman 1972; 1991: 71; Allinger-Csollich
1998: 316–19) depicts the six lowest stages of a ziggurat. Wiseman showed that the top of the
tablet, broken oˆ, most likely showed a seventh level whose dimensions formed a cube.
5. An unprovenanced fragmentary Babylonian tablet in Berlin (VAT 8322 + 12886), of uncertain
date, gives a drawing of six concentric squares. Jakob-Rost (1984) read the inscription on the
upper margin of the tablet as the “ziggurat of Marduk,” suggesting that the assumed temple on
top of the structure counted as the seventh stage and wondering if the plan could have repre-
sented the ziggurat of Borsippa or an earlier form of the one at Babylon. More recently, George
(2005/2006: 76) has disputed this reading, though allowing that the diagram is “in any case,
clearly schematic” and may have represented at least an idealized ziggurat.
6. As seen, the name of the ziggurat at Borsippa (é.ur4.(me).imin.an.ki) may imply a sevenfold
structure.
7. A fragmentary tablet from Nippur now kept in Jena (HS 200a; Oelsner 1984, 1989) gives the
˘oorplan of a building in the form of seven concentric squares with proportionally decreasing
dimensions. The inscription does not identify the edi˜ce, but Suter (1997: 5–6) convincingly
argues that it can hardly have been intended for anything other than a ziggurat. While Oelsner
thought the palaeography may suggest the Old Babylonian or Kassite period, Suter (1997: 6) has
proposed an Ur III date.
8. The ziggurat of Istar at Uruk was called (é).gi6.pàr.imin.(na/bi), “house of seven giparu’s”
(George 1993: 93 s.v. #384), giparu meaning the residence of the high priestess (Weadock 1975).
A passage in the Temple Hymns (16.201–202) arguably refers to this structure when it charac-
terizes Eanna as é.ub.imin, “house with seven corners” (ETCSL 4.80.1) or “of seven niches”
(George 1993: 154 s.v. #1154). In practice, it is hard to conceive how else a building with seven
“corners,” “niches,” or “angles” (Hübner and Reizammer 1985: 1092–93 s.v. “ub I,” “ub II,” “ub III,”
“ub IV’) could have been realized architecturally other than in the form of seven stages, unless
it were a heptagon (as translated in Edzard 1997: 143).
9. A byname given by Gudea, ruler of Lagas (ca. 2100 BC), to the é.PA of Ningirsu, either at Girsu
or at Lagas, was é.ub.imin, “house of seven niches” (George 1993: 94 s.v. #393; 154 s.v. #1154,
#1155) or “of seven corners” (Sjöberg and Bergmann 1969: 91–92), the same name as that given
to the Eanna at Uruk (see #8). It has long been thought that the é.PA was a ziggurat (Barton
1923, but see Falkenstein 1996: 132–34).
The cumulative eˆect of this evidence is di¯cult to brush aside. It would appear that, possibly as early as
the late-third millennium BC, the ziggurat in optima forma consisted of seven tiers. Even if, in some cases,
the original intention had been to build ziggurats of seven stages, this may not always have been pos-
sible for architectural and economic reasons, in the same way that European cathedrals have not
always been brought to completion (see George 2005–2006: 86). As a further analogy, hundreds of
medieval churches were carefully aligned to the cardinal points, but the fact that just as many were not
does not disprove the former. Similarly, solar alignments were built into the layout of Stonehenge, but
while this may suggest that one of its functions was a Sun temple, this hardly applies by default to
every circle of standing stones in prehistoric Britain. It only demonstrates diˆerences in the priorities
of the builders. Diˆerent Mesopotamian cities were “owned” by distinct gods, and there may have been
complex theological and as yet unfathomed reasons for building towers with diˆering numbers of stages.
Short

ZIGGURATS, COLORS, AND PLANETS: RAWLINSON REVISITED 65
Ziggurats and Colors
The next step is to reevaluate the possible association of the tiers of the ziggurat to colors. To what
extent might a uniform pattern of colors have been at work in the design and decoration of ziggurats,
particularly those erected or refurbished in the ˜rst millennium BC? At Khorsabad, Place and Thomas
(allowed by Frankfort) found the sequence white – black – red – blue. This matches exactly that at Ur,
if one includes the white courtyard as Woolley did: “The four colours in their order, white, black, red,
and blue” (1925: 14). The same sequence may also have underlain the arrangement at Borsippa, where,
despite other uncertainties, we know that the base was black and that blue-glazed bricks once decorated
a higher stage (see above).
Though limited, the evidence raises the possibility that a consensus color scheme was employed in the
decoration of ziggurats during the ˜rst millennium BC—at least with respect to the relative order of the
colors white, black, red, and blue. This is, of course, the very sequence given by Herodotus for the four
outermost “battlements” of Ecbatana, reviving the question of whether his account may, after all, have
re˘ected a blueprint from the Neo-Babylonian period. With regard to the remaining colors in the Ec-
batana sequence (light red, silver, and gold), one should not be too skeptical given that Place and Thomas
reported the discovery of glazed tiles around the Khorsabad ruin sporting these very colors. Further,
the restriction of silver and gold decoration to the topmost stages is credible enough, for purely practical
considerations. As it happens, silver and gold are mentioned in characters drawn at the center of the
seven-staged ziggurat plan on the Jena tablet (Suter 1997: 5). Instead of Suter’s hunch that these signify
some sort of treasure preserved in the building, they could have been the intended colors of the top
two levels of the ziggurat, if only to a token degree.
The alleged color sequence discerned by Rawlinson at Borsippa ˘atly contradicted that given by
Herodotus, with the result that he was tempted to “correct” the tradition regarding Ecbatana. A hundred
and ˜fty years of further archaeological research have produced a very telling result. We have seen
that Rawlinson’s color scheme at Borsippa was largely imaginary. By contrast, the Ecbatana tradition
has been partly con˜rmed by the evidence from other sites (Khorsabad and Ur). A reexamination of the
possible meaning of this color sequence (hereafter “the Ecbatana/ziggurat sequence”), especially where
it has been con˜rmed by excavation, is thus long overdue.
Aside from the placement of precious metals on the topmost levels and the protective function of the
bitumen at the base, it is hard to imagine any practical reason for the positioning of the other (better-
attested) colors, namely, white, red, and blue. All the more since they appear to have occurred in a ˜xed
sequence. One understandably turns, therefore, to the likelihood that their arrangement had some religious
or cosmological signi˜cance. But did this concern a sequence of planets, as Rawlinson had assumed?
Intermezzo: Rawlinson’s Sources
As Graeco-Roman astronomers have long been known to have linked the respective planets to speci˜c
colors (see conveniently Boll 1916: 20)—primarily based on their appearance—Rawlinson’s assump-
tion, in his day, that the Babylonians matched colors to planets, is in general perfectly reasonable. Yet to
what extent can his sources be said to be representative of Babylonian astronomy?
Rawlinson’s statement that “Herodotus’ story [regarding Ecbatana] was a fable of Sabaean origin”
(1841: 127) implies anachronistically that the Sabaeans, a gnostic sect of northern Syria, predated
Herodotus. In addition, and by his own admission, Rawlinson relied almost exclusively on the twelfth-
century AD poet Nizami for an alleged Sabaean tradition, and even this source he did not follow consis-
tently. Speci˜cally, Rawlinson elected to assign the colors yellow and silver to Venus and the Moon
(1861: 21), instead of Nizami’s white and green (Wilson 1924).

PETER JAMES AND MARINUS ANTHONY VAN DER SLUIJS66
There is a more fundamental problem with Rawlinson’s reliance on Nizami. Rawlinson may not
have been unjusti˜ed in his belief that the Sabaeans of Harran had inherited a measure of ancient
Mesopotamian planet lore; their eclectic religion comprised a mixture of hermeticism, Jewish gnos-
ticism, Neo-Platonism, and indigenous Assyrian beliefs (Green 1992). But was Nizami a reliable mouth-
piece of Sabaean traditions? It seems unlikely. Nizami’s Haft Paikar was not a work on astrology, but
a romance and, although contemporary with the Sabaeans, Nizami’s background was Azerbaijani.
A far more reliable, though later, witness to the Sabaeans would be the Arab geographer al-Dimasqÿ
(† 1327 AD), who gave a detailed description of the pagan temples of Harran, including the statues,
colors, and metals associated with each of the seven planet gods venerated there (Nuhbat al-Dahr,
1.10.1–8; 2.4.1, trans. Mehren 1874: 41–47; 71; see also Chwolsohn 1856: 381–96).16 His writings,
together with those of the tenth-century AD scholar al-Nihawandÿ (apud Ibn al-»awzÿ, in Hjärpe 1972:
69–88), reveal the following sequence of correspondences between planets, colors, and metals:
Thus, Rawlinson’s alleged “Sabaean” colors compare very poorly with those known from more representa-
tive sources. There are only four matches (Sun, Moon, Saturn, Mars), but there is complete disagree-
ment on Jupiter (Rawlinson: red; Harran: green), Venus (Rawlinson: yellow; Harran: blue/white), and
Mercury (Rawlinson: blue; Harran: brown). It transpires that Rawlinson’s much-vaunted “Sabaean”
system was not Sabaean at all. Rawlinson’s cavalier approach to the archaeology of Birs Nimr¿d was
thus compounded by use of inappropriate source material.
In addition, when reconstructing the colors at Borsippa, Rawlinson invoked “the well known
[planetary] order of Saturn, Jupiter, Mars, Sol, Venus, Mercury, and the Moon. . . .” But well known to
whom? This was in fact the “Ptolemaic” sequence prevailing in late Hellenistic and Roman times,
which was often referred to as the “Chaldaean” order of planets (Evans 1998: 348–49; Neugebauer
1975: 691). The placement of the Sun in the center of the sequence re˘ects a quasi-heliocentric system,
which the consensus of scholarship would not date any earlier than the second century BC, certainly
not in a Babylonian context. The Rawlinsons never addressed this chronological di¯culty, believing
that the Babylonians
. . . had notions not far from the truth with respect to the relative distance from the earth of the sun, moon, and
planets . . . probably based upon a knowledge, more or less exact, of the periodic times which the several bodies
occupy in their (real or apparent) revolutions. From the diˆerence in the times the Babylonians assumed a corre-
sponding diˆerence in the size of the orbits, and consequently a greater or less distance from the common centre.
(Rawlinson 1879: 577)
Yet the only evidence oˆered for such claims, bordering on circularity, was that “the arrangement of
the great temple at Borsippa . . . is a su¯cient proof. . . .” The Pan-Babylonianists, who followed the
Rawlinsons’ leads in many respects, are generally guilty of having back-projected Hellenistic concepts
into earlier times (Parpola 2004: 239).
Counting Seven Planets
Despite the weakness of Rawlinson’s case, his core idea of linking a seven-tiered ziggurat with the
notion of seven planets is still worthy of examination. While it is often overlooked, it is clear that Baby-
16. Long before Rawlinson, al-Dimasqÿ (1.10.11) curiously attributed the Sabaean belief system to the “Chaldaeans.”
Saturn Jupiter Mars Sun Venus Mercury Moon
black green red gold blue / white brown silver / white
lead tin iron gold copper mercury silver
One Line Long

ZIGGURATS, COLORS, AND PLANETS: RAWLINSON REVISITED 67
lonian astronomers did conceptualize a grouping of seven planets (Horowitz 1998: 153; Reiner 1995: 4).
Although their astronomical texts tend to exclude the Sun and the Moon from lists of planets, a number,
ranging from the MUL.APIN to the Seleucid era, group all seven bodies together.
MUL.APIN (ca. 1000 BC, but see Geller 1990) states that the Sun, Jupiter, Venus, Mars, Mercury and
Saturn, respectively, all travel “the (same) path the Moon travels,” concluding with the words: “Together
six gods who have the same positions, (and) who touch the stars of the sky and keep changing their
positions” (II i 1–8, trans. Hunger and Pingree 1989: 70–71).17 Thus as Horowitz (1998: 172) observes,
“all seven ancient planets were to be found along a single band of the sky.” The so-called Great Star
List (Neo-Assyrian; ll. 242–244, trans. Koch-Westenholz 1995: 200–201) enumerates: “The moon and
the sun Jupiter Venus / ‘Wild Sheep’: Saturn Mercury / Mars” and concludes with the summary state-
ment “Seven planets.”18 The seven bodies are also listed together on two mid-seventh-century tablets
from Nineveh, both of which list Moon – Sun – Jupiter – Venus – Saturn – Mercury – Mars (K. 2067,
19–21, in Weidner 1915: 19–20; K. 4386 = An-ta-gál, G. 303–309, ed. Cavigneaux et al. 1985: 229).19
The date of these attestations, unlike those in Seleucid texts (e.g., AO 6460 obv. 22–24, trans. Sachs
1969: 338; Rochberg-Halton 1988, 324–25, 328), precludes Greek in˘uence.
Matching Planets and Colors
In testing a possible link between ziggurat colors and planets, an obvious advantage today is that
speci˜c associations between planets and colors can now be inferred directly from excavated texts
unavailable in Rawlinson’s time. While Neo-Babylonian omen texts generally feature the planets in a
variety of colors, depending on atmospheric and other conditions (e.g., for Venus; Reiner and Pingree
1998: 19), at least one text (K 2346 54) gives what appears to be a “standard” list of planetary colors:
The white star is Jupiter, the Red Star is Mars, the Green star is Venus, the Black star is Saturn, variant: Mercury.
(Reiner and Pingree 1998: 249; Brown 2000: 143)20
For three of these planets, the respective colors given are readily con˜rmed in other sources (Reiner and
Pingree 1998: 248–49; Eilers 1976: 79–80, 93). A common Babylonian name for Jupiter was mulbabbar,
“white star,”21 Mars, conspicuous for its redness, was known as mulSA5, “red star,” and Saturn, widely asso-
ciated with black in the ancient world, is described in Babylonian texts as mulgíg, “dark star” or mulMI
(see Gössmann 1950: 28).22 The color associations of Venus and Mercury require further discussion.
As Sumerian and Akkadian made no distinction between green and blue (Landsberger 1967: esp.
139), the notion of a “green” Venus is arguably con˜rmed by the intimate association of the goddess of
Venus, Inanna-Istar, with lapis lazuli, a blue mineral that was valued in the cults of the gods generally,
17. The names used for these bodies are Sin, †amas, †ulpaea, Dilibat, Salbatanu, †ihtu sa Ninurta, and Kajamanu.
18. The respective terms are d30 u, dutu, dsul.pa.è, mul dil.bat, mul udu.idim, mul sag.us, dudu.idim.gu4.ud, and
mul Sal-bat-a-nu, in all “7 mu l udu.idim.mes.”
19. The names used in An-ta-gál are dA.KU = dsin (30), d kas-se-biGUD.LUGAL = dUTU, dda-pi-nu = dsul-pa-è-a, d zibZIG =
ddil-bat, dlu-lim = dUDU.IDIM.SAG.U†, dbi-ib-bu = dUDU.IDIM.GU4.UD, and dsi-mu-ud = dSal-bat-a-nu.
20. The Sumerian terms are MUL BABBAR, “white star,” MUL SA5 “red star,” MUL SIG7, “green star” and MUL MI, “black
star” or “dark star.”
21. Plato (Republic, 617a) alludes to Jupiter as the “whitest” (leukótatos) of the seven planets and to Venus as deúteron . . .
leukóteti, the “second whitest.” Despite that, white has been assumed by many (from Rawlinson to Parpola) to correspond to Venus,
even though there seems to be no Babylonian source ascribing this color to Venus, with the exception of one di¯cult omen text in
which it is described as “white and black”; Reiner and Pingree propose that this simply refers to the planet being bright or ob-
scured (1998: 19).
22. Similarly in India, Saturn was known as Kala, “the black one” (Eilers 1976: 93). Maimonides (Commentary on the Misna,
‘Abodah Zarah, 3.1, in Chwolsohn 1856: 485), too, described Saturn as black. See Ptolemy, Tetrabiblos 2.9; Nigidius Figulus, apud
Lucan, Pharsalia 1.651–657; Celsus, apud Origen, Contra Celsum, 6.22.

PETER JAMES AND MARINUS ANTHONY VAN DER SLUIJS68
but among the planetary deities especially so with Istar. Many references can be found to support this.
For example, in Inanna’s Descent to the Netherworld, when the goddess prepares for her descent she
grasps a measuring rod and line of lapis lazuli and dons a necklace of two lapis lazuli beads (ll. 107, 113).
When appealing to the gods Enlil, Nanna, and Enki to save her life, she repeatedly refers to herself as
the stone: “Don’t let your precious lapis lazuli be split there with the mason’s stone.” (ll. 45, see also 54,
62, 187, 201, 214; trans. ETCSL 1.4.1) In the so-called Uruk text, Inanna is given the epithet “her of the
lapis lazuli (gems)”; a date-gatherer presents her with lapis lazuli from the “gem-revealing heap” under-
neath a date-palm (Jacobsen 1976: 34–35). Most importantly, a Kassite-period list of magical correspon-
dences from Nippur, of which a partial Neo-Assyrian copy survives, directly links lapis lazuli (uqnû)
with the planet Venus (ddili.bat).23 Indeed, a blue/green color for the planet, whatever its rationale,24
is attested in at least one classical source25 and ethnographic sources from some other cultures with no
demonstrable link to Mesopotamia.26
Mercury is associated with a red hue, in at least one Babylonian omen text: “If Venus wears a red
crown—Mercury stands in front of her” (K. 148.13; Reiner and Pingree 1998: 58–59; Brown 2000: 60).27
Finally, the association of gold and silver with the Sun and Moon is universal and requires no further
comment.
To sum up, the following planet-color associations may be viewed as standard for Neo-Babylonian
times: Jupiter (white), Mars (red), Venus (green/blue), Saturn (black), Mercury (red), Sun (gold), Moon
(silver).
The Ziggurat Planet Order?
While the backlash against the Pan-Babylonianists has receded, allowing astral interpretations of
Mesopotamian religion and myth to be discussed afresh, this time with more restraint (e.g., Heimpel
1986), discussion of a connection between the ziggurats and planetary astronomy has remained in the
doldrums. It required the boldness of an eminent Assyriologist to raise the matter in recent years.
Stressing the archaeological evidence, Parpola has revived the idea of a planetary meaning for the colors
of the Khorsabad ziggurat and Ecbatana battlements:
Remains of colouring on the ziggurat of the Assyrian capital city Dur-†arruken show that each of its stages was
painted in a diˆerent colour, the sequence of colours corresponding to the colouring of the seven concentric walls
of Ecbatana in Herodotus I 98 (white, black, purple, blue, orange, gold, silver) and probably symbolizing the seven
planetary spheres (Venus, Saturn, Mars, Mercury, Jupiter, Sun, and Moon). (1997: XCII; 2000: 199)
23. Livingstone (1986: 175–87, esp. 176–77 ). The collation (involving ˜ve manuscripts) given by Livingstone supersedes the
incomplete version published by Langdon (1919: 330–43).
24. Venus easily shifts to the blue to the unaided eye. Livingstone (1986: 182) remarked: “Similarity in appearance may underlie
connection of lapis lazuli with Venus.”
25. Pseudo-Callisthenes, Life of Alexander, 1.8 (Armenian and Syriac versions).
26. For example, one of the names assigned to Venus as evening star by the Mescalero Apache of New Mexico is su≥u≥s bik?
edalatl? izhe?, “star blue/green under it”; apparently, this “gives emphasis to the characteristic glow of color Apaches perceive to
be a part of Venus.” (Farrer 1986: 60). The Mexican Annals of Cuauhtitlan (Codex Chimalpopoca, 3.56–4.2) report that the god
Quetzalcoatl “was placed in his mother’s belly when she swallowed a piece of jade” (trans. Bierhorst 1992: 28). The same god was
thought to have metamorphosed into Venus’s aspect as morning star, in the guise of a quetzal bird. (Codex Chimalpopoca, 7.27–
46.) The quetzal bird (Pharomachrus mocino) has red feathers on the chest, but green wings and tail feathers.
27. In this text, the various colors of Venus’s “crown” are ascribed to the in˘uence of occultating planets. The cuneiform for
“red” is SA5. This is the same term as that used for Mars and in several texts of the same period also Jupiter (although the latter
is never called “the red planet”), so in this text at least there is no distinction between “dark red” and “light red.” The connection
of Mercury with red may relate to Plato’s (Republic 617a) quali˜cation of Mercury as xanthóteros, the comparative of xanthós,
“yellow, of various shades, freq. with a tinge of red, brown, auburn . . .” (LSJ 1187 s.v. “xanqovÍ”; cf. the brown of the Sabaeans,
above). Vettius Valens associated Mercury with ochrós (Boll 1916: 20), “pale, wan, of complexion . . . esp. pale-yellow, sallow . . .”
(LSJ 2042 s.v. “w˚crovÍ”).
One Line Long

ZIGGURATS, COLORS, AND PLANETS: RAWLINSON REVISITED 69
Unfortunately, the detail of this proposal is subject to the objection raised by Jensen to Rawlinson’s
theory in 1890 (above): the deduced sequence of planets is unrecognizable either in astronomical
terms or in the planetary orders familiar from Assyro-Babylonian texts. But this is because Jensen (and
Parpola, too) relied on the manifestly unreliable planet-color correspondences oˆered by Rawlinson.
Would a more intelligible order emerge if the planet-color associations given in cuneiform sources
were applied to the Ecbatana/ziggurat sequence? The experimental result would be as follows:
A signi˜cant improvement on Rawlinson, this analysis relies on eighth- to seventh-century BC cuneiform
sources contemporary with the construction of the ˜rst-millennium ziggurats. Intriguingly, the sequence
reconstructed above produces a meaningful result in astronomical terms, as its constituents neatly fall
into three groups: lowest are the outer planets (Jupiter, Saturn, and Mars); then come the inner planets
(Venus and Mercury); and uppermost are the two major luminaries (Moon and Sun). The position of the
latter is perhaps supported by the single or double pair of horns reported to surmount some ziggurats.28
These possibly represent the crescent of the moon god Sîn, typically envisioned as a bull,29 or the sun
god †amas, who could also be referred to as a bull.30
With respect to the ˜ve traditional planets, their order is astronomically “correct” from a geocentric
perspective, except for the reversal of the two outermost planets, Saturn and Jupiter. Yet it can hardly
be coincidence that this order including the inversion is identical to that given in the ˜rst/second
century AD by pseudo-Eratosthenes (Catasterismi, 43) and in the late-˜rst century by Hyginus (Poetica
Astronomica, 2.42).31 In switching the positions of Saturn and Jupiter, as compared to the standard
systems in vogue in Greece from the fourth century BC onwards, this order is aberrant and possibly
archaic.
The astronomical ceilings known from Egyptian tombs from the mid- to late-second millennium BC
reveal a sequence—like that reconstructed from the Ecbatana/ziggurat color order—which similarly
distinguishes the outer from the inner planets and features the same reversal of Saturn and Jupiter,
leaving Mars as the lowest of the outer planets:
The usual order in which the planets appear on the monuments prior to the Graeco-Roman period is Jupiter,
Saturn, Mars, Mercury and Venus, with the ˜rst three, the outer planets, separated from the last two, the inner
planets, by . . . triangle decans . . . (Neugebauer and Parker 1969: 175, also 3; see also Clagett 1995: 124)
The Egyptian New Kingdom evidence thus refutes the possible objection that Babylonian astronomers
of the eighth and seventh centuries BC could not yet have conceived a planetary order re˘ecting fairly
accurate knowledge. The question of possible exchange of knowledge between Mesopotamia and Egypt
on these matters will require further study.
white black dark red blue light red silver gold
Jupiter Saturn Mars Venus Mercury Moon Sun
28. “Bovine horns are depicted at the tops of ziggurats and other buildings throughout the ancient Near East . . .” (Horowitz
1998: 124) The locus classicus is En¿ma Elis VI.66, where with respect to the ziggurat built by Marduk, it states: “To the base of
Esharra its horns look down” (Speiser 1969: 69; Horowitz 1998: 123). See also the reference in Assurbanipal’s “annals,” eighth
campaign: “I destroyed the ziggurat of Susa, which was made of blue bricks. I cut oˆ its horns made of shining cast copper . . .”
(Horowitz 1998: 124). A relief of Assurbanipal depicts the same ziggurat, formed of ˜ve tiers with two pairs of horns on top
(Parrot 1949: 47–48).
29. E.g., Sîn’s epithet “the red wild bull” (Eilers 1976: 32).
30. E.g., Lugalbanda and Enmerkar (ll. 223–24), where Lugalbanda addresses the Sun as “Bright bull, emerging from heaven’s
base” (Heimpel 1986: 143). In the Gilgamesh Epic (IV, Ha1 10–13; trans. George 1999: 37) Enkidu interprets the “wild bull” of a
dream as †amas.
31. Some manuscripts of Hyginus also swap the names Phaenon and Phaethon, so that the usual Hellenistic assignment of
Phaethon to Jupiter and of Phaenon to Saturn is restored, but Phaethon/Jupiter occupies the highest orbit (Le Boeu˙e 1983: 178 n. 7).

PETER JAMES AND MARINUS ANTHONY VAN DER SLUIJS70
The Neo-Babylonian Planet Order
A more serious objection to the hypothetical planet order reconstructed on the basis of ziggurat
decoration would be that it con˘icts with the standard sequence of the ˜ve “proper” planets known from
Neo-Babylonian astronomical texts (Brown 2000: 143; Rochberg-Halton 1988: 323, 327–28; Neugebauer
1975: 690):32
Jupiter – Venus – Saturn – Mercury – Mars
The con˘ict here may only be apparent, however. As it happens, the Ecbatana ziggurat order can be
correlated with the standard Neo-Babylonian sequence by means of a pentagram,33 by plotting either
on the corners and reading along the diagonal lines (˜g. 1).34 Such a result is not a mere curiosity. It is
well known that, in imperial Roman times, the sequence of the weekdays (by their tutelary deities) was
derived from the quasi-heliocentric (“Chaldaean”) planet order then in vogue (Cassius Dio 37.18–19;
Boll 1911: 372–75; Neugebauer 1957: 169; 1975: 691) by taking leaps of two. While the precise origins
of this weekday system—still adhered to—remain shrouded in fog (Zerubavel 1985: 8–9; Sarton 1959:
32. The same sequence survived in Greek horia astrology (Rochberg-Halton 1988: 323). Variant orders are seen in MUL.APIN
(II i 1–6: Sun – Jupiter – Venus – Mars – Mercury – Saturn, with the Moon) and during the Seleucid period (Jupiter – Venus –
Mercury – Saturn – Mars; e.g., TCL VI.41, 23–24, trans. Sachs 1969, 338; see Neugebauer 1975: 690; Rochberg-Halton 1988: 323;
Brown 2000: 143).
33. As a symbol, the pentagram is known from the Uruk and Jemdet Nasr periods (Falkenstein 1936: 118–19 s.v. #453; Goˆ
1963: 77, 113). Intriguingly, the pentagram was the Sumerian sign for u b, ˜rst attested in the Uruk IV period and used for the
“regions” or “corners” of the world (De Vogel 1966: 292). While, in relation to cardinal directions, these were most frequently
counted as four, the names of some ziggurats testify to a notion of “seven corners” as well (see above). De Vogel concludes: “It is
certain that the pentagram comes from Babylon. . . . The pentagram, perhaps originally a geometrical ˜gure, acquired a cosmic
signi˜cance as early as in the third millennium B.C. (the connotation of heavenly region or quarter, already in the Fara texts, say
c. 2600), while somewhat later it was connected with the planets” (1966: 292, 296). A heptagram shown on an intriguing late
Babylonian tablet from Nippur was evidently used as a mathematical device of some sort (Horowitz 2006). Waerzeggers and Siebes
have suggested it depicts the seven strings of the harp, while stressing “that the present interpretation of CBS 1766 rests on a
number of emendations to the edition by W. Horowitz that could not yet be con˜rmed by collations” (2007: 45). Yet, even if the tablet
concerns musical tuning, it is worth asking if this precludes an astronomical connotation, as originally considered by Hilprecht
(1903: 530).
34. This approach was explored by Tetens (in Jensen 1886: 267; 1890: 133), Winckler (1905: 192–93), and Boll (1911: 373–75).
Fig. 1. Planetary Pentagram.

ZIGGURATS, COLORS, AND PLANETS: RAWLINSON REVISITED 71
326–33; Gandz 1948–1949: 215–16; Colson 1926: esp. 6, 18, 22), there is no doubt that it originated as a
mathematical adaptation of the so-called Chaldaean order.
35
Signi˜cantly, one way in which this adap-
tation could conveniently be visualized was the use of a heptagram (˜g. 2) as a graphic device (Winckler
1905: 192; Jeremias 1911: 43; Sarton 1959: 332–33). Although none of the extant textual sources spells
this out, Hellenistic astrologers are known to have positioned the planets at the respective corners of
a hexagon, with Bel or Jupiter in the center—in Palmyra, Zaghuan (Tunisia), and possibly Baalbek,
where “the hexagonal court becomes a symbol of the days of the week and the month. . . .” (Brown 1939;
Palmyrene image in Wood 1753: pl. XIX A).
The quasi-heliocentric planetary system underlying the weekday order must, of course, have been
known to the Babylonian astronomers of the Hellenistic (Seleucid) period, if only because a full-blown
heliocentric (Aristarchean) model had been advocated by the Babylonian astronomer Seleucus in the
second century
BC
(Sarton 1959: 295; Plutarch,
Platonic Questions
, 8.1 [1006C]). Yet neither this nor
the earlier Platonic-Aristotelian geocentric order is attested in
any
extant cuneiform text. Instead, texts
from this period continued to place Jupiter and Venus at the front of the sequence, as had been customary
in the Neo-Babylonian period, while the relative positions of Mercury, Mars, and Saturn varied (Rochberg-
Halton 1988: 325, 328). As Rochberg-Halton (1988: 323) notes, this positioning . . .
has nothing to do with a spatial arrangement of the planets in the cosmos, in contrast to the following planetary
arrangement in Greek astronomy and astrology: . The Greek model represents the order of the planets in
35. Zerubavel (1985: 14) quali˜ed “the astrological seven-day week” as “essentially a Hellenistic invention,” which “most
probably evolved sometime during the second century B.C. at the very heart of the Hellenistic world, namely Alexandria,” almost
concurrently with the Ptolemaic sequence itself.
Fig. 2. Planetary Heptagram.

PETER JAMES AND MARINUS ANTHONY VAN DER SLUIJS72
depth according to their periods of sidereal rotation. No such “natural” explanation can be oˆered for the Babylo-
nian sequence.
As Rochberg-Halton explains, the order typically followed in Seleucid literature is an astrological
one, determined by the bene˜c and male˜c aspects of the respective planets (1988: 323–25; see also
Neugebauer 1957: 169). This is amply illustrated by a Seleucid text that gives the following order of
planets, with accompanying glosses: Jupiter (favorable)—Venus (calm)—Mercury (heroic)—Mars (am-
biguous)—Saturn (dark, disturbed). The reasons behind these astrological associations are beyond the
scope of the present study—though it must be signi˜cant that Jupiter and Venus are the two brightest
planets, Saturn the dimmest—except to note that their origins well predate the Seleucid period.36
The astrological planetary order was used in all literary genres, not just the astronomical.37 That a
scienti˜c order is unattested in Seleucid cuneiform texts, when astronomers must have been thoroughly
aware of developments in “Greek” astronomy, suggests that it was proscribed for magico-religious
reasons. In plain terms, a “taboo” appears to have been operating. One is reminded here of the secrecy of
Egyptian and “Chaldaean” astronomers stressed by Strabo (Geog. 17.1.29). According to him, Greek access
to “barbarian” knowledge of the planets was a protracted process. At ˜rst the information was deliberately
withheld, rather “concealed,” by their savants, and only gradually revealed to those (like Eudoxus during
his long sojourn in Egypt) who had the patience to “prevail upon them in time and by courting their
favour,” This report is consistent with the so-called Geheimwissen formula attached to some Babylonian
astrological tablets, according to which “knowledge of the contents of the tablet is reserved for the ini-
tiated, lit. ‘the one who knows,’ and that showing it to an uninitiated person is a crime as bad as destroying
the tablet itself. . . . In a sense, all astrology, including the series En¿ma Anu Enlil, was certainly con-
sidered a part of the secrets of the great gods” (Koch-Westenholz 1995: 95). As the colophon of one tablet
states, “Reading of (what has to do with) the great gods is the secret lore of Heaven and Earth. Reading
the commentary is the secret lore of the scholars” (Ki 1904-10-9, 94: 26–30, in Koch-Westenholz 1995: 96).
These considerations require us to reexamine the problem of the standard Neo-Babylonian order of
the planets, which, like the Seleucid, conspicuously evades any “natural” explanation in terms of astro-
nomical distance. As both the Neo-Babylonian and the Seleucid “auspicious” orders begin with Jupiter
and Venus, one must accept Rochberg-Halton’s opinion that the same astrological thinking (that of bene˜c
and male˜c planets) underlay both schemes. Given this, as in the case of the Seleucids, one should be wary
of denying Neo-Babylonian scholars any interest in a diˆerent, scienti˜c order of the planets—though
such may not have been “written” it may have been encoded in the color decoration of ziggurats.
In conclusion, while direct evidence is lacking for the planetary signi˜cance of the Ecbatana/ziggurat
color sequence as reconstructed here, three lines of evidence strongly support it: (a) that it is meaningful
in astronomical terms, (b) that it is similar to the planet order attested in some Greek and Latin writings
and sources from the Egyptian New Kingdom, and (c) that the device of the pentagram apparently
links it to the known standard planetary order of Neo-Babylonian times. There may be thus be good
reason to reappraise the persistent tradition of the Greeks that they learnt about the planets from the
Babylonians and Egyptians.38 This tradition con˘icts with the current understanding that the Greeks
“invented” the concept of a planetary order based on relative distance, while the Babylonian view of
36. As Rochberg-Halton (1988: 328) observed, “the bare traces [of the schema] . . . are already found in apodoses of En¿ma
Anu Enlil and its Old Babylonian forerunners.” See Rochberg-Halton (1988: 326) for the example of Mars.
37. “Regardless of textual genre, therefore, the planets were enumerated by convention. In my view, that convention
was originally shaped by the underlying astrological schema identifying planets as either bene˜c or male˜c” (Rochberg-Halton
1988: 328).
38. See, e.g., Philip of Opus, Epinomis 986D–987D; Aristotle, Cael. 2.12 (292a8–9); Diodorus Siculus 2.30; Seneca, Nat. 7.3.1–2;
Proclus, In Platonis Timaeum Commentarius, III.125.28–30.

ZIGGURATS, COLORS, AND PLANETS: RAWLINSON REVISITED 73
the stars was a purely arithmetical one, with the stars and planets all existing together on one undiˆer-
entiated plane (Vernant 1983: 176–89; Rochberg-Halton 1988: 323; Lloyd 1991: 294–95; Pedersen
1993: 63; Evans 1998: 320).
From “Heavens” to “Planetary Heavens”
The case of the Neo-Babylonian ziggurats strongly suggests the concept of a sevenfold strati˜cation
of the cosmos. Does any independent evidence support such a concept? In the literature on ziggurats, a
fact that has often been overlooked is that Babylonian texts attest to a strati˜ed view of the heavens.
While some texts describe the heavens as having three layers, respectively the Upper, Middle, and
Lower Heavens (Horowitz 1998: 8–15), many others refer to “seven heavens” or “seven heavens and
seven earths” (Horowitz 1998: 208–20, 244). The evidence largely comes from Sumerian incantations,
known at least from the Old Babylonian period onwards. The motif is apparently re˘ected in later
(Jewish and Arab) traditions concerning seven superimposed ˜rmaments and underworlds.
It would seem that the term “seven heavens and seven earths” relates to the layers of the sky and
their infernal counterparts, respectively. Like the layered sky, the underworld was conceived as having
seven zones, each guarded by its gateway and gatekeeper, as known from Inanna’s Descent to the Nether-
world and Nergal and Ereskigal (contra Horowitz 1998: 358).
The concept of “seven heavens” and “seven earths” could well have found its architectural expres-
sion in the seven-tiered ziggurats. In the annual Akÿtu ritual, Marduk was symbolically imprisoned in
Etemenanki, as explicitly stated in Marduk’s Ordeal (VAT 9555+9538; ND 812(a), l. 13; Livingstone
1986: 236–37, also 212–13, l. 40): “[That which] they do [on] the ziggurat: Because the gods imprisoned
him he disappeared and was held captive inside.” He was released from captivity by Nabû (ll. 8–9), the
tutelary deity of astronomy to whom the Borsippa ziggurat was dedicated. This, in combination with
Strabo’s reference (16.1.5) to the ziggurat of Babylon as the Belou táphos, the “tomb of Belus [Marduk],”
indicated to many scholars, though not all, that Marduk’s enacted imprisonment can be seen in terms of
a ritual death scheme, suggesting that the ziggurat acted as a representation of the underworld (Pallis
1926: 102–10, 243; Frankfort 1948: 322–23; Livingstone 1986: 230–31; Naydler 2005: 56; more generally,
Busink 1938: 48–49; but see Bidmead 2002: 87).39
That there was an infernal component to the symbolism of ziggurats may also be inferred from the
use of the Sumerian word ku r both for a mountain and for the underworld (e.g., Katz 2003: 110), as
ziggurats were often conceived as forms of a cosmic mountain (Busink 1938: 48–49; Walton 1995: 159–
62). Indeed, in Marduk’s Ordeal (ll. 6, 7), the place where the messenger comes to ˜nd Marduk is de-
scribed as the hur-sa-an or “mountain” (Pallis 1926: 221–22, 242).40
The question remains whether and when ziggurats, as a model of a sevenfold cosmos, may have
become connected with the planets.41 As seen, analysis of the ziggurat/Ecbatana sequence indicates
that by Neo-Babylonian times the planets could have been associated with seven-tiered ziggurats. The
39. In keeping with this, Parpola opined that the seven stages of a ziggurat symbolized the seven gates of the underworld
(2000: 199) and that “through its seven-staged colouring the ziggurat is associated both with the rainbow and the descent and ascent
of Istar” (1997: XCII). If Inanna’s ziggurat at Uruk did indeed have seven stages, as suggested above (#8 in the list of seven-staged
ziggurats), this raises the possibility that the narrative of the goddess’s descent, in which she is gradually divested of garments
and ornaments referred to as me, accompanied a ritual performed on this ziggurat.
40. Livingstone’s translation “the place of the river ordeal” is less literal (1986: 237). Gelb et al. (1965: 253–54), s.v. “hursanu
A (hursanu),” give hursanu as “mountain.”
41. If the color scheme of ziggurats symbolized a planetary order, it remains as yet unclear why the outermost planets would
have been associated with the bottom and not the top of the structure. At the very least, such an arrangement would conveniently
have allowed the topmost stages, decorated with the most costly ornaments, to be dedicated to the two major luminaries.

PETER JAMES AND MARINUS ANTHONY VAN DER SLUIJS74
respective stages of the ziggurat may not have been associated with individual planets from the
start, but a tendency to do just that may well have developed in subsequent times, culminating in a
˜rst-millennium trend of ziggurats with seven “planetary” colors.
It would be simplistic to argue that ˜rst-millennium ziggurats were built mainly or even largely to
represent the seven planets. Yet even if this was not the primary function of ziggurats, some may have
acquired such symbolism over time, as a “bonus” coming on top of the more fundamental associations
such as those with the cosmic mountain or the underworld.
Conclusions
We have come a long way from our starting point, which was to reassess the meaning of Herodotus’s
description of the colored “battlements” built by Deioces at Median Ecbatana about 700 BC. It was
Herodotus’s account that inspired Henry Rawlinson to investigate the possibility that the speci˜ed colors
spelled out a sequence of planets and that this could provide a major clue to ancient astronomical knowl-
edge. It was a potentially brilliant insight. Yet when he applied it to Borsippa, Rawlinson eˆectively de-
molished his own case by an over-imaginative interpretation of the archaeology and an inappropriate
use of a color sequence from an allegedly “Sabaean” tradition, leading him to an absurd position where
he had to throw into doubt the order given by Herodotus. Had more cuneiform sources been available
to him, he may well have come to a diˆerent conclusion, that is, that a Neo-Babylonian color-planet
order matching that of Ecbatana was applied to ziggurat architecture.
Our reinvestigation has led to three key insights, which are valid independently of each other, namely,
that the Assyro-Babylonians did conceive of seven-tiered ziggurats; that the stages of some ziggurats
were decorated in diˆerent colors; and that a particular symbolical color was assigned to each of the
planets. These three threads may have converged into a single system during the ˜rst millennium BC.
Without any doubt, a very similar planet order to that reconstructed here for the Neo-Babylonian
period existed much earlier in Egypt, in which the outer and inner planets were correctly distinguished.
The implications for the development of Greek knowledge of the solar system are considerable, restor-
ing con˜dence in the Greek claim that their knowledge of the planets derived from the ancient Near
East. Such questions—in a post-Pan-Babylonianist phase of scholarship—need serious reexamination. The
Babylonians and Egyptians seem to have shared a model of a strati˜ed cosmos that involved a rudi-
mentary concept of planetary distance and, at the very least, the distinction between inner and outer
planets. This may have represented a large part of the knowledge gift for which Greek thinkers felt so
indebted. The Greek contribution to these matters would not, then, have been the perception of the uni-
verse in geometric terms per se (planetary distance), but the introduction of circular geometry in terms of
planetary orbits, to which the names of Pythagoras and Parmenides have traditionally been attached.42
Once introduced, the idea of circular orbits (rather than simple strati˜cation) led to the growth of plan-
etary astronomy proper.
The resolution of a “minor” puzzle of archaeoastronomy, namely, Rawlinson’s interpretation of the
Ecbatana colors, may thus have wider rami˜cations for the history of astronomy than one might expect.
Acknowledgments
This article could not have been completed without the invaluable support of the Mainwaring
Archive Foundation and the Caeno Foundation. We are also extremely grateful to Simo Parpola, Mark
42. Diogenes Laertius, Life of Pythagoras, 25–26, 48; Life of Parmenides, 21; see also Theo of Smyrna, Mathematics, 150.18;
Van der Waerden (1979: 445, 453). For a discussion of the pioneers in the related concept of a spherical cosmos, see Kahn (1970).

ZIGGURATS, COLORS, AND PLANETS: RAWLINSON REVISITED 75
Geller, Claudia Suter, John Bimson, Ev Cochrane, Nick Thorpe, and an anonymous JCS referee for
reading drafts and providing comments, extra references and encouragement; to Wilfrid Allinger-
Csollich, Andrew George, Alan Gri¯ths, Baruch Halpern, Hermann Hunger, Joachim Marzahn, and
Wilfred van Soldt for providing information and answering questions; and to Seong Hee Jo for help
with the illustrations.
References
Allinger-Csollich, W.
1998 Birs Nimrud II. “Tieftempel”—“Hochtempel”: Vergleichende Studien Babylon—Borsippa. BaM 29:
95–330.
Averbeck, R. E.
2000 The Cylinders of Gudea (2. 155). Pp. 417–33 in The Context of Scripture, Vol. 2: Monumental Inscrip-
tions from the Biblical World, eds. W. W. Hallo and K. L. Younger. Leiden: Brill.
Babelon, E.
1906 Manual of Oriental Antiquities. London: Grevel.
Barton, G. A.
1923 The Form and Nature of E-Pa at Lagash. JAOS 43: 92–95.
Beaulieu, P. A.
2000 Nebuchadnezzar II’s Restoration of E-Urimin-Ankia, The Ziggurat of Borsippa (2.122b). Pp. 309–10 in
The Context of Scripture, Vol. 2: Monumental Inscriptions from the Biblical World, eds. W. W. Hallo and
K. L. Younger. Leiden: Brill.
Bidmead, J.
2002 The Akÿtu Festival; Religious Continuity and Royal Legitimation in Mesopotamia. Gorgias Dissertations:
Near Eastern Studies 2. Piscataway, NJ: Gorgias.
Bierhorst, J.
1992 History and Mythology of the Aztecs: The Codex Chimalpopoca. Tucson: The University of Arizona Press.
Boll, F.
1911 Zur babylonischen Planetenordnung. Zeitschrift für Assyriologie und verwandte Gebiete 25: 372–76.
1916 Antike Beobachtungen farbiger Sterne. Abhandlungen der Königlichen Bayerischen Akademie der
Wissenschaften: Philosophisch-philologische und historische Klasse 30.1. Munich: Königlich Bayerischen
Akademie der Wissenschaften.
Brown, D.
2000 Mesopotamian Planetary Astronomy-Astrology. CM 18. Groningen: Styx.
Brown, D. F.
1939 The Hexagonal Court at Baalbek. AJA 43: 285–88.
Budge, E. A. W.
1884 Babylonian Life and History. By-Paths of Bible Knowledge 5. London: R. T. S.
Busink, Th. A.
1938 De Toren van Babel: Zijn Vorm en zijn Beteekenis. Batavia Centrum: Noordhoˆ-Kolˆ.
Cavigneaux, A.; Güterbock, H. G.; and Roth, M. T., eds.
1985 The Series Erim-hus = anantu and An-ta-gál = saqû. MSL 17. Rome: Ponti˜cium Institutum Biblicum.
Chwolsohn, D.
1856 Die Ssabier und der Ssabismus II. St. Petersburg: Buchdruckerei der kaiserlichen Akademie der Wissen-
schaften.
Clagett, M.
1995 Ancient Egyptian Science, A Source Book: Calendars, Clocks, and Astronomy II. Philadelphia: American
Philosophical Society.
Colson, F. H.
1926 The Week: An Essay on the Origin and Development of the Seven-Day Cycle. Cambridge: Cambridge
University Press.
De Vogel, C. J.
1966 Pythagoras and Early Pythagoreanism: An Interpretation of Neglected Evidence on the Philosopher
Pythagoras. Assen: Van Gorcum.

PETER JAMES AND MARINUS ANTHONY VAN DER SLUIJS76
Drews, R.
1973 The Greek Accounts of Eastern History. Cambridge, MA: Harvard University Press.
Edzard, D. O.
1997 Gudea and His Dynasty. RIME 3/1. Toronto: University of Toronto Press.
Eilers, W.
1976 Sinn und Herkunft der Planetennamen. SBAW 5. Munich: Verlag der bayerischen Akademie der
Wissenschaften.
Evans, J.
1998 The History and Practice of Ancient Astronomy. Oxford: Oxford University Press.
Falkenstein, A.
1936 Archaische Texte aus Uruk. Ausgrabungen der Deutschen Forschungsgemeinschaft in Uruk-Warka 2.
Berlin: Deutsche Forschungsgemeinschaft.
1996 Die Inschriften Gudeas von Lagas I. Analecta Orientalia; Commentationes Scienti˜cae de Rebus Orientis
Antiqui 30. Rome: Ponti˜cium Institutum Biblicum.
Farrer, C. R.
1986 Mescalero Apache Terminology for Venus. Archaeoastronomy 9: 59–61.
Frankfort, H.
1948 Kingship and the Gods. Chicago: University of Chicago Press.
1970 The Art and Architecture of the Ancient Orient. 4th edition. Harmondsworth: Penguin.
Gandz, S.
1948–
1949
The Origin of the Planetary Week or the Planetary Week in Hebrew Literature. American Academy for
Jewish Research; Proceedings 18: 213–54.
Gelb, I. J.; Jacobsen, Th.; Landsberger, B.; and Oppenheim, A. L., eds.
1956 The Assyrian Dictionary of the Oriental Institute of the University of Chicago, Vol. 6. Chicago: The
Oriental Institute of the University of Chicago.
Geller, M.
1990 Astronomy and Authorship. BSOAS 53: 209–13.
George, A. R.
1993 House Most High: The Temples of Ancient Mesopotamia MC 5. Winona Lake, IN: Eisenbrauns.
1999 The Epic of Gilgamesh: The Babylonian Epic Poem and Other Texts in Akkadian and Sumerian.
Harmondsworth: Allen Lane/The Penguin Press.
2005–
2006
The Tower of Babel: Archaeology, History and Cuneiform Texts. AfOr 51: 75–95.
Goˆ, B. L.
1963 Symbols of Prehistoric Mesopotamia. New Haven: Yale University Press.
Golzio, K.-H.
1983 Der Tempel im alten Mesopotamien und seine Parallelen in Indien: Eine religionshistorische Studie.
Beihefte der Zeitschrift für Religions- und Geistesgeschichte 25. Leiden: Brill.
Gössmann, P. F.
1950 Planetarium Babylonicum oder die sumerisch-babylonische Sternnamen. Rome: Verlag des päpstlichen
Bibelinstituts.
Green, T. M.
1992 The City of the Moon God: Religious Traditions of Harran. Religions in the Graeco-Roman World 114.
Leiden: Brill.
Grene, D.
1987 The History of Herodotus. Chicago: University of Chicago Press.
Heimpel, W.
1982 A Catalog of Near Eastern Venus Deities. Syro-Mesopotamian Studies 4: 59–72.
1986 The Sun at Night and the Doors of Heaven in Babylonian Texts. JCS 38: 127–51
Henige, D.
2004 Herodotus’ Median Chronology from a Slightly Diˆerent Perspective. Iranica Antiqua 39: 239–48.
Hilprecht, H. V.
1903 Exploration in Bible Lands during the 19th Century. Edinburgh: T&T Clark.
Hjärpe, J.
1972 Analyse Critique des Traditions Arabes sur les Sabéens Harraniens. Uppsala: Skriv.

ZIGGURATS, COLORS, AND PLANETS: RAWLINSON REVISITED 77
Horowitz, W.
1998 Mesopotamian Cosmic Geography. Winona Lake, IN: Eisenbrauns.
2006 A Late Babylonian Tablet with Concentric Circles from the University Museum (CBS 1766). JANES 30:
37–53.
How, W. W., and Wells, J.
1912 A Commentary on Herodotus 1. Oxford: Clarendon.
Hübner, B., and Reizammer, A., eds.
1985 Inim Kiengi II; Sumerisch-Deutsches Glossar 2. Marktredwitz: H. Troeger.
Hunger, H., and Pingree, D.
1989 MUL.APIN: An Astronomical Compendium in Cuneiform. AfO Beiheft 24. Horn: F. Berger.
Jacobsen, T.
1976 The Treasures of Darkness: A History of Mesopotamian Religion. New Haven: Yale University Press.
Jakob-Rost, L.
1984 Zur Zikkurrat von Babylon. Forschungen und Berichte; Staatliche Museen zu Berlin 24: 59–62.
Jastrow, M.
1898 The Religion of Babylonia and Assyria. Boston: Ginn.
Jensen, P.
1886 Der Kakkab mísrîå der Antares. Zeitschrift für Assyriologie und verwandte Gebiete 1: 244–67.
1890 Die Kosmologie der Babylonier: Studien und Materialien. Strassburg: Trübner.
Jeremias, A.
1911 The Old Testament in the Light of the Ancient East; Manual of Biblical Archæology. Theological Trans-
lation Library 28. London: Williams & Norgate.
Kahn, J. C.
1970 On Early Greek Astronomy. JHS 90: 99–116.
Katz, D.
2003 The Image of the Netherworld in the Sumerian Sources. Bethesda, MD: CDL.
Koch-Westenholz, U.
1995 Mesopotamian Astrology: An Introduction to Babylonian and Assyrian Celestial Divination. CNIP 19.
Copenhagen: Museum Tusculanum Press.
Koldewey, R.
1911 Ausgrabungen der Deutschen Orient-Gesellschaft in Babylon; I: Die Tempel von Babylon und Borsippa.
WVDOG 15. Leipzig: Hinrichs.
Kugler, F. X.
1907 Sternkunde und Sterndienst in Babel; assyriologische, astronomische und astralmythologische Unter-
suchungen I–II. Münster: Asschendorˆsche Verlagsbuchhandlung.
Landsberger, B.
1967 Über Farben im Sumerisch-Akkadischen. JCS 21: 139–73.
Langdon, S.
1919 Sumerian Liturgies and Psalms. Publications of the Babylonian Section X. 4. Philadelphia: The Uni-
versity Museum.
Le Boeu˙e, A.
1983 Hygin: L’Astronomie. Paris: Les Belles Lettres.
Lewis, C. T., and Short, C., eds.
1879 A Latin Dictionary. Oxford: Clarendon.
Livingstone, A.
1986 Mystical and Mythological Explanatory Works of Assyrian and Babylonian Scholars. Oxford: Clarendon.
Lloyd, G. E. R.
1991 Methods and Problems in Greek Science. Cambridge: Cambridge University Press.
Mehren, A. F.
1874 Manuel de la cosmographie du moyen age, traduit de l’arabe “Nokhbet ed-dahr ˜ ’Adjaib-il-birr wal-
bah’r” de Shems ed-Dÿn Abou-’Abdallah Moh’ammed de Damas. Copenhagen: Reitzel.
Naydler, J.
2005 Shamanic Wisdom in the Pyramid Texts; The Mystical Tradition of Ancient Egypt. Rochester, VT: Inner
Traditions.

PETER JAMES AND MARINUS ANTHONY VAN DER SLUIJS78
Neugebauer, O.
1957 The Exact Sciences in Antiquity. 2nd ed. Providence, RI: Brown University Press.
1975 A History of Ancient Mathematical Astronomy II. Berlin: Springer.
Neugebauer, O., and Parker, R. A.
1969 Egyptian Astronomical Texts III. Decans, Planets, Constellations and Zodiacs. Brown Egyptological
Studies 6. Providence, RI: Brown University Press.
Oelsner, J.
1984 Ein Zikkurrat-Grundriss aus Nippur. Forschungen und Berichte; Staatliche Museen zu Berlin 24: 63–65.
1989 Ein Zikkurrat-Grundriß aus Nippur—Addendum. Forschungen und Berichte; Staatliche Museen zu Berlin
27: 51.
Oppert, J.
1863 Expédition Scienti˜que en Mésopotamie Exécutée par Ordre du Gouvernement de 1851 a 1854 par
MM. Fulgence Fresnel, Félix Thomas et Jules Oppert 1. Paris: Imprimerie Impériale.
Pallis, S. A.
1926 The Babylonian Akitu Festival. Historisk-˜lologiske Meddelelser XII,1. Copenhagen: K. Dansk Vidern-
skabernes Selskab.
Parpola, S.
1997 Assyrian Prophecies. SAA 9. Helsinki: Helsinki University Press.
2000 Monotheism in Ancient Assyria. Pp. 165–209 in One God or Many? Concepts of Divinity in the Ancient
World, ed. B. N. Porter. Transactions of the Casco Bay Assyriological Institute 1. Chebeague Island, ME.
2004 Back to Delitzsch and Jeremias: The Relevance of the Pan-Babylonian School to the MELAMMU Project.
Pp. 237–47 in MELAMMU Symposia IV, eds. A. Panaino and A. Piras. Milan: Università di Bologna &
Istituto Italiano per l’Africa e l’Oriente.
Parrot, A.
1949 Ziggurats et Tour de Babel. Paris: Albin Michel.
Pedersen, O.
1993 Early Physics and Astronomy: A Historical Introduction. Cambridge: Cambridge University Press.
Place, V., and Thomas, F.
1870 Ninive et l’Assyrie 2. Paris: Imprimerie Impériale.
Potts, D. T.
1999 The Archaeology of Elam. Cambridge: Cambridge University Press.
Rawlinson, G.
1858 The History of Herodotus. London: John Murray.
1879 The Five Great Monarchies of the Ancient Eastern World II. London: John Murray.
Rawlinson, H. C.
1841 Memoir on the Site of the Atropatenian Ecbatana. Journal of the Royal Geographical Society of London
X:I: 65–158.
1861 On the Birs Nimrud, or the Great Temple of Borsippa. JRAS 18: 1–34.
Reiner, E.
1995 Astral Magic in Babylonia. Transactions of the American Philosophical Society Held at Philadelphia for
Promoting Useful Knowledge 85:4. Philadelphia: The American Philosophical Society.
Reiner, E., and Pingree, D., eds.
1998 Babylonian Planetary Omens. Part Three. CM 11. Groningen: Styx.
Rochberg-Halton, F.
1988 Bene˜c and Male˜c Planets in Babylonian Astrology. Pp. 323–28 in A Scienti˜c Humanist: Studies in
Memory of Abraham Sachs, eds. E. Leichty, M. deJ. Ellis and P. Gerardi. Occasional Publications of the
Samuel Noah Kramer Fund 9. Philadelphia: The University Museum.
Sachs, A.
1969 Akkadian Rituals. Pp. 331–45 in Ancient Near Eastern Texts Relating to the Old Testament, ed. J. B.
Pritchard. Princeton: Princeton University Press.
Sarton, G.
1959 A History of Science 2. Cambridge, MA: Harvard University Press.
Sayce, A.
1887 Lectures on the Origin and Growth of Religion as Illustrated by the Religion of the Ancient Baby-
lonians. Hibbert Lectures. London: Williams and Norgate.

ZIGGURATS, COLORS, AND PLANETS: RAWLINSON REVISITED 79
Schmidt, W.
1972 The Origin and Growth of Religion: Facts and Theories. New York: Cooper Square.
Scurlock, J. A.
1990 Herodotos’ Median Chronology Again?! (sun “including” or “excluding”). Iranica Antiqua 25: 149–63.
Singor, H. W.
1992 The Achaean Wall and the Seven Gates of Thebes. Hermes 120: 401–11
Sjöberg, Å. W., and Bergmann, E.
1969 The Collection of the Sumerian Temple Hymns. TCS III. Locust Valley, NY: Augustin.
Speiser, E. A.
1969 Akkadian Myths and Epics. Pp. 60–119 in Ancient Near Eastern Texts Relating to the Old Testament, ed.
J. B. Pritchard. Princeton: Princeton University Press.
Suter, C. E.
1997 Gudeas vermeintliche Segnungen des Eninnu. ZA 87: 1–10.
Szarzynska, K.
1993 Oˆerings for the Goddess Inana in Archaic Uruk. RA 87: 7–26.
Van der Waerden, B. L.
1979 Die Pythagoreer: Religiöse Bruderschaft und Schule der Wissenschaft. Die Bibliothek der Alten Welt.
Zürich: Artemis.
Vernant, J. P.
1983 Myth and Thought among the Greeks. London: Routledge/Kegan Paul.
Waerzeggers, C., and Siebes, R.
2007 An Alternative Interpretation of the Seven-Pointed Star on CBS 1766. NABU 2007/2: 43–45.
Walton, J. H.
1995 The Mesopotamian Background of the Tower of Babel Account and its Implications. Bulletin for Biblical
Research 5: 155–75.
Weadock, P. N.
1975 The GIPARU at Ur. Iraq 37: 101–28.
Weidner, E. F.
1915 Handbuch der babylonischen Astronomie, 1: Der babylonische Fixsternhimmel; Beiträge zur ältesten
Geschichte der Sternbilder. AB 23. Leipzig: Hinrich.
Wilson, C. E.
1924 The Haft Paikar (The Seven Beauties): Containing the Life and Adventures of King Bahram Gur, and
the Seven Stories Told him by his Seven Queens / By Nizami of Ganja; Translated from the Persian, with
a Commentary. London: A. Probsthain.
Wilson, D.
1851 The Archaeology and Prehistoric Annals of Scotland. Edinburgh: Sutherland and Knox.
Winckler, H. F.
1905 Altorientalische Forschungen iii. 2. Leipzig: Eduard Pfeiˆer.
Wiseman, D. J.
1972 A Babylonian Architect? AnSt 22: 141–47.
1991 Nebuchadrezzar and Babylon. The Schweich Lectures of the British Academy. Oxford: Oxford University
Press.
Wood, R.
1753 The Ruins of Palmyra otherwise Tedmor, in the Desart. London.
Woolley, C. L.
1925 The Excavations at Ur, 1923–1924. The Antiquaries Journal V.1: 9–15.
1939 Ur Excavations, Vol. V: The Ziggurat and Its Surroundings. Publications of the Joint Expedition of the
British Museum and of the University Museum, University of Pennsylvania, Philadelphia to Meso-
potamia. London: Oxford University Press.
1954 Excavations at Ur: A Record of Twelve Years’ Work. London: Ernest Benn.
Zerubavel, E.
1985 The Seven Day Circle; The History and Meaning of the Week. New York: The Free Press.