PDF hosted at the Radboud Repository of the Radboud University

Nijmegen

The following full text is a publisher's version.

For additional information about this publication click this link.

http://hdl.handle.net/2066/105757

Please be advised that this information was generated on 2017-07-04 and may be subject to
change.
BABesch 77 (2002)

The Nemrud Dağ Project: first interim report

Eric M. Moormann and Miguel John Versluys
with contributions by Maurice Crijns, Onno van Nijf & Marlies Schipperheijn,
Tesse D. Stek & Ellen Thiermann and Anne ten Brink

1 Introduction 4.1.1 Goals

2 Previous research on Nemrud Dağ 4.1.2 How to use the SIS
3 A short description of the monument 4.1.3 Block drawings
3.1 The East Terrace 4.2 Some new observations
3.1.1 Lay-out and architectural elements 4.2.1 Cosmological orientation?
3.1.2 The colossal statues 4.2.2 The Lion horoscope: proposal for a new
3.1.3 The stelae and sculptures in tuffit dating (Maurice Crijns)
3.1.4 Ancestor stelae and altars 4.2.3 Chronology: a new dating frame?
3.2 The West Terrace 4.2.4 The lower ensemble on the East Terrace
3.2.1 Lay-out and architectural elements 4.2.5 The ‘Fire Altar’ on the East Terrace: true
3.2.2 The colossal statues or false
3.2.3 The stelae and sculptures in tuffit 4.2.6 The West Terrace in its oldest shape
3.2.4 Ancestor stelae and altars 4.2.7 The sequence of the colossal statues
3.3 The North Terrace 4.2.8 Style and iconography
3.4 Building materials 4.3 The pilot survey (with Anne ten Brink)
3.5 Preservation, construction and technique 4.3.1 Introduction
3.5.1 The stelae 4.3.2 Results
3.5.2 The colossal statues 4.3.3 Conclusion
3.5.3 The nomos inscription and other texts 4.4 Epigraphic research (Onno van Nijf &
4 The Amsterdam University research Marlies Schipperheijn)
4.1 The SIS (with Tesse Stek & Ellen 5 Outlook for the next years
Thiermann) Bibliography/Abbreviations

Fig. 1. Kommagene (after Wagner 2000).

73
1 INTRODUCTION 14th of July 2001, the same day as represented on
the famous Lion horoscope on the West Terrace
The International Nemrud Foundation (hereafter (see pp. 97-99). The team was composed of
INF), established in 1998, has worked for a cou- archaeologists, epigraphists, geodesists, construc-
ple of years to obtain a permit from the Turkish tion and AutoCAD engineers.1
Government to carry out archaeological research The 2001 research had several goals. In view of
on and around the monument of Antiochos I of the project’s overall goals of preservation and
Kommagene on the mountain of Nemrud Daǧ in restoration most energy was spent to obtain a
the province of Adıyaman, Southeastern Anatolia, documentation as complete as possible, the so-
Turkey. The main aim of the Foundation is to pro- called SIS (Site Information System), which may
tect the monument from further deterioration and serve to both students and restorers in the future.
destruction as well as to reconstruct the colossal Archaeologists could profit from this system for
statue groups and some of the relief stelae on the their research. At first impression such a basic
East and West Terraces. In view of the ambitious documentation of the site might seem redundant,
aims of the project it was deemed necessary to as an impressive monograph of the site was pub-
make a new documentation of the remains on the lished in 1996 presenting the results of the
mountain and to study the possibility of a virtual American excavations directed by Theresa B.
reconstruction. Goell in the 1950’s, 1960’s and 1970’s.2 However,
After a permit had been granted by the Turkish Goell’s results and documentation proved to be
Ministry of Culture to the University of Amsterdam, problematic in several ways. Besides, we expect
a team from the Amsterdam Archaeological that new technology (especially the laser scan)
Centre of this university started research on the could yield a more fundamental documentation

Fig. 2. Map of ancient Kommagene showing the most important archaeological sites (after Wagner 2000).

74
of the site than hitherto provided. We had, there- by Karl Humann and Puchstein in 1883 and one
fore, to review critically the work of Goell for the Ottoman exploration by Osman Hamdi Bey and
compilation of our SIS; our impressions will be Osgan Effendi in 1883 made the monument
published over the next years with the intention famous to the world of archaeology at once. The
to serve as a kind of review article of Sanders two teams made thorough descriptions of the vis-
1996.3 ible remains and concentrated on the collection of
Moreover, though concentrating on the terraces the epigraphic material, especially the large
and statues proper, the Amsterdam research also inscription in Greek on the backs of the colossi
made a modest start with the study of the direct that informs us about the monument, its use and
context of the monument, i.e. the area of the moun- the intentions of its founder (the nomos inscrip-
tain below the tumulus and terraces. Therefore, we tion, figs. 26-27). Although they do not provide a
surveyed a part of this area, first of all to see if synthesis, their work remains fundamental.
there were any stray finds that could provide new It took some decades before more thorough in-
information for understanding the monument. vestigations started. The German scholar Friedrich
There is, for instance, a remarkable absence of Karl Dörner wrote his PhD on Kommagene and
pottery from the monument itself that could have travelled to the area from 1939 onwards. He was
been preserved on the mountain’s slopes. More- to become the discoverer or explorer of the main
over, we know that this area was incorporated in other sites in Kommagene such as the ancestor
the conception of the monument by means of grave and royal summer residence Arsameia at the
guarding animals and entrance stelae. A second Nymphaios (discovered by Dörner in 1953) and
aim was to look if a more extensive field survey the tumuli – hierothesia – of Karakuş and Arsameia
is feasible. at the Euphrates.6 Between 1953 and 1973, the
It has often been noted that despite the unique American archaeologist Theresa B. Goell con-
character of the monument, and the fact that it is ducted a large number of campaigns on Nemrud
one of Turkey’s most important archaeological Dağ.7 Both scholars made restorations and recon-
monuments, our understanding of the site, and structions of various elements of this complex.
the state of our documentation are relatively poor. The primary aim of Goell’s first campaigns was
The low interest shown in the site by Classical to clear the terraces form the debris, in order to
and Oriental archaeologists, may well be due to understand the organisation of space. Members
the fact that it blends Graeco-Roman and Persian of her team worked on aspects like the statuary
elements, as was suggested by Goell. However, it remains and all fragments of sculpture and epig-
would seem that Goell herself was partly respon- raphy found during the clearance. The main out-
sible also, on account of her long delay in pub- come of Goell’s work is that the West and East
lishing the results of her research. For these rea- Terraces were pendants, that contained more or
sons we felt obliged to provide a detailed account less the same elements. Afterwards, her work
of our intentions, working methods and results focused more and more on the discovery of the
on a yearly basis. The results and insights of the tomb of Antiochos, mentioned in the inscription
campaigns will thus be published each following at the back of the statues, Antiochos’ nomos.
year in this periodical. These interim reports, The pioniering efforts of Dörner and Goell gen-
offering work in progress and provisionary con- erated a series of other studies on Nemrud Daǧ,
clusions, will form the base of our final publica- usually from a historical or epigraphic point of
tion. Obviously, only at that time final conclusions view; such as the studies by Helmut Waldmann
will be formulated. and Sencer Şahin.8 An important contribution to
the debate was made by Jörg Wagner who made
2 PREVIOUS RESEARCH ON NEMRUD DAĞ the whole region of Kommagene, its history and
its archaeological monuments from various peri-
The remains on Nemrud Dağ, a hill majestically ods to his field of study, thereby broadening the
overlooking the anti-Taurus, were discovered in background against which a monument like
1881 by the German engineer Karl Sester.4 Until Nemrud Dağ should be studied (fig. 2).9
that time Kommagene was no more than a name The German and Turkish pioneers of the 1880’s
known from historical sources only, studied by were only able to document the visible remains,
the Leiden classicist David Jacobus van Lennep which they did very well considering the cir-
in his 1828 PhD and by Theodor Mommsen in an cumstances. They lacked the means to remove the
article of 1876 (fig. 1).5 Two short German expe- thick layers of gravel from the remains which had
ditions by Sester and Otto Puchstein in 1882 and rolled down from the tumulus that crowns the

75
 Fig. 3. Nemrud Dağ: topographical map of the monument (© INF).

mountain. Their maps, therefore show big blank 3 A SHORT DESCRIPTION OF THE MONUMENT
spots. Dörner did not carry out excavations, but
published many important observations. Goell, The monument - named hierothesion in the large
on the other hand, turned every stone in her inscription on the back of the colossi - consists of
quest to find the ‘truth’ about the monument.10 a tumulus flanked at three sides by terraces. The
She has worked longer on the mountain than any tumulus ‘erected’ on top of Nemrud Dağ suggests
other scholar, but unfortunately she only pub- the presence of the tomb of Antiochos as men-
lished short reports, and popular articles, that tioned in the inscription. It is formed of the orig-
contained little information on the scope of her inal mountain, partly topped and hewn off on its
project, her working methods, or the result of the surface and a covering layer of gravel. Small
project. Her many notes and drafts were collected square terraces were constructed to hold the
and published posthumously by Donald Sanders gravel.11 The height of the tumulus is about 50
in 1996 (Sanders 1996). In fact this important vol- metres (fig. 3).
ume forms the basis for further investigations of The workers started with cutting off the north-
the monument. west flank between the North and the West
Before presenting the Amsterdam research and Terraces. Subsequently the gravel that was pro-
offering some new hypotheses concerning the his- duced this way was transported to ‘deposits’, viz.
tory and interpretation of the monument, we give the slopes between the two terraces mentioned.
a short description of it. Then they constructed artificial terraces and

76
 Fig. 4. Nemrud Dağ,
plan of the East Terrace (© INF).

started chopping the top. The gravel was spread text as descriptive as possible, largely drawing on
out towards the lower layers and was partly kept the information provided by the work of Goell
in place by the artificial banks.12 The deposits and our own observations in the field. Our criti-
were used to backfill the empty spaces. The part cal discussion of several aspects and some new
of the mountain still standing at the south side of hypothesis can be found in paragraph 4.2.
the West Terrace was not entirely used for the
gravel production. It was, however, gradually 3.1 The East Terrace
topped of as can be seen from cuttings in the rock.
Maybe some kind of platform was constructed. 3.1.1 Lay-out and architectural elements
The large amount of gravel still visible at the
south side is an extension of the tumulus. The East Terrace is a rectangular area that has
As has become clear, mainly from Tom Utecht’s been flattened by cutting off part of the mountain.
research, the tomb is probably not located in the At the west side it is bordered by a stepped area
tumulus itself but either under it or at some place cut out from the rock. On top of this, on a second
in the vicinity.13 At four sides Goell made holes plateau largely also hewn out from the mountain,
into the mountain to discover the tomb. The the colossal statues are erected. On the east side
results of explosions with dynamite are clearly border, opposite to this ensemble, Goell has recon-
observable behind the colossi on the East and the structed a stepped structure in tuffit; the so-called
West Terraces. fire altar. She did so on the basis of the presence
In the description of the terraces, we will each of stepped structures at the north and south and
time first describe the setting itself, before shortly a retaining wall at the east, already seen by
introducing the architectural elements. Matters of Humann and Puchstein (see further paragraph
specific interest deserving more attention, like, for 4.2.5). At the south side a steep mountain slope
instance, the colossal statues will be treated in begins; to the north there is a path leading to the
separate paragraphs. We have tried to keep the North Terrace (fig. 4).

77
 Fig. 5. Nemrud Dağ, East Terrace, overview (2001, photo E.M. Moormann).

At the south side of the terrace there is an square angles used; as already remarked by
entrance: eleven steps (of which nine are pre- Wolfram Hoepfner who suggests that a certain
served) have been cut out in the living rock. The ‘Naturnähe’ was looked for (fig. 5).14
rest of the stair makes five turnings of 90o as one
can observe in the cutting of the rock downwards 3.1.2 The colossal statues
the slope. The precise form cannot be recon-
structed unless by voiding the space from the Main feature of the East Terrace is a row of nine
masses of gravel. A small platform, in line with colossal statues: five statues of (semi-)gods
the eleven steps, seems to command the progres- flanked at both sides by a lion/eagle pair. The
sion of the stair. The main entrance to the terrace podium on which the colossal statues stand has
seems to have been at the north-east side: a so- two steps in front, hewn out from the rock. At the
called procession road, hiera hodos, ended just north there are three steps, probably to be com-
below the terrace at this side. The existence of this bined with those found by Goell at the terrace
road is certain by the find of an inscribed entrance itself and leading to the reconstruction of stairs.
stele on the slope, besides one can still see part of The extreme south and north sides of this podium
the path hewn out from the rock. have been damaged and the statues of the lion and
Bases for stelae and corresponding altars in the eagle at the south and that of the lion at the
front of them line the north and south borders of north have completely tumbled down (figs. 5-6).
the terrace (see further paragraph 3.1.4). At the back of the statues a path of 3.5 m width
Goell reported the presence of various tuffit has been spared out. By this way the nomos,
animals on the terrace. She reconstructed the inscribed on the lower blocks of the statues, could
presence of two pairs of an eagle and a lion on or be reached. The path runs behind the statues
at the so-called fire altar. Nowadays a lion (186 along the west side, not only giving access to those
cm H) standing next to the stepped structure and who wanted to read the inscriptions on the backs
fragments of two eagles along the slope rest from of the statues, but maybe also used for other pur-
this adornment. The passway to the North Terrace poses in the rituals (fig. 7).
would also have been decorated with a large eagle The largest figure, Zeus-Oromasdes, stands in
of tuffit. No remains of this are visible anymore. the centre. He is flanked by Apollo-Mithras-Helios-
Lastly, it must be noted that the whole setting Hermes and Artagnes-Herakles on his left and
is irregular and that there are no precise axes and Kommagene and Antiochos on his right (for this

78
 Fig. 6. Nemrud Dağ, East Terrace, the colossal statues (2001, photo T.D. Stek).

Fig. 7. Nemrud Dağ, East Terrace, the back of the colossal statues with the path (2001, photo E.M.
Moormann).

79
sequence of the statues cf. further section 4.2.7).
The different statues are named in the nomos
inscription.15 At both ends a standing eagle and a
squatting lion have been placed.
The figures are seated on huge thrones, their
feet resting on footstools standing between the
throne’s legs. They do not show any movement;
Zeus and Kommagene only have the right foot set
forward. The statue of Zeus protrudes some 21 cm
in respects to the other figures. The pairs of ani-
mals share their base and are worked in the
round. The colossi have a basis of their own,
formed by the feet of the throne and the footstool
in front.16 The backs are summarily rendered. The
lower three tiers, bearing the inscription, are plain
flat walls. The upper layers have rounded out-
lines and show no details. The coarseness of the
modelling of the bodies contrasts to the precise
and detailed working of the heads (figs. 8-9).
The five figures of (semi-)gods are composed
of seven or eight horizontal tiers of limestone
blocks, from bottom to top: 1) footstool; 2) feet Fig. 8. Nemrud Dağ, East Terrace, statue of Zeus,
and legs until lower edge of tunic; 3) legs up to upper part as standing (2001, photo T.D. Stek).
knees clad in long garment; 4) lap and forearms;
5) breast and upper arms; 6) shoulders; 7) head;
8) top gear. Tiers 6-8 are monolithic. The eighth
level includes Kommagene’s kalathos, the tips of
the tiaras of Zeus and Apollo, whereas Herakles
lacks such an extra layer. The other tiers can be
variously composed of two or more blocks.
The five lower layers of the colossi are still
intact. The figure of Kommagene has, apart from
her kalathos lying in front of the lower plateau,
been preserved completely, until it came down,
according to Goell probably by lighting, between
1963 and 1967. The head and fragments of the back
now lie behind the statues. Here is also standing
the head of Antiochos. The heads of the other fig-
ures and blocks belonging to the animals are scat-
tered over the East Terrace.
The dress of the figures is barely detailed and
shows few folds apart from the shoulders where
the tips of the mantle have been put together with
a buckle. The male dress is oriental: boots, trousers,
long-sleeved tunic and cloak. Kommagene has the
Greek chiton and himation, of which some folds
are indicated near the knees. The mouths are
parted and the eyes are suggested to look upward,
showing a certain degree of (Hellenistic) pathos.
The male gods’ tunics end below the knees and a
tip of the mantle is indicated by one fold at the
sides of the feet. Other folds can be seen in the lap
and (diagonally) on the breast. Kommagene’s Fig. 9. Nemrud Dağ, East Terrace, statue of Zeus,
long dress falls down between the legs and its head lying on the terrace (2001, photo E.M.
drapery fills this space. The feet are shod in sum- Moormann).

80
 Fig. 10. Nemrud Dağ, East Terrace, the statues of Zeus (partly, left), Apollo, Herakles and Eagle
(2001, photo T.D. Stek).

marily worked boots with laces and lappets.

Kommagene has her hair parted in the middle
and combed backwards. Her ears are adorned
with long earrings. She is crowned with a thick
wreath composed of corn poppies in the centre
and fruits, mainly pomegranates and grapes to
the sides. A veil covers the back of the head and
a kalathos, now standing on the terrace, was once
on top of the head.
Antiochos, Zeus and Apollo have a bundle of
weed, the so-called barsom in their left hand on
their laps.17 Herakles sports his club with his left
hand against his shoulder and Kommagene does
the same with the cornucopia. Her right hand
holds a bunch of fruit in her lap.
The heads look brighter of colour than the rest
of the figures (this colour is also predominant on
the West Terrace), which may be due to the dif-
ferent processes of weathering. The heads of
Herakles, Kommagene and the northern eagle
show numerous small cavities in the surface
caused by algae.

Fig. 11. Nemrud Dağ, East Terrace, statue of Apollo,

head standing on the terrace (2001, photo E.M.
Moormann).

81
 Fig. 12. Nemrud Dağ, East Terrace, the statues of Antiochos (left) and Kommagene (2001, photo T.D. Stek).

The two eagles are represented standing erect,

with the large feet with five toes protruding over
the base. Feathers are not indicated and the birds’
surface is entirely smooth. The heads look grim
with - as it were - a frown created by a bulging
piece of flesh over the inner corners of the well-
articulated almond-shaped eyes. The lions are
seated on their hind legs; the tail is curled at the
outside of the row, over the paw of these legs. The
forepaws show four toes and are adorned with a
metal ring (fig. 13). The mane is carefully worked
on the breast between the legs and on the head
and back. Both heads are badly damaged but look
similar to those preserved on the West Terrace:
rather friendly, with open fangs, tongue jutting
out and upper row of teeth visible.

3.1.3 The stelae and sculptures in tuffit

A podium has been cut out of the rock in front of

Fig. 13. Nemrud Dağ, East Terrace, the forepaws of the statues. In front there are tuffit steps made
the southern side Lion (2001, photo E.M. Moormann). from blocks placed against the prepared rock.

82
Goell reconstructed three steps. Sanders however,
in his examination of Goell’s evidence, gives the
number of two steps and proposes good argu-
ments to support his view.18 Goell reported that
fragments of large stelae similar to those on the
West Terrace have been found. She offers a recon-
struction of a Lion horoscope and of the four dex-
iosis stelae in front of the statues on one of the
tuffit steps (see section 3.2.3). At the southern side
of the podium four sockets for stelae have been
hewn out in the rock; the third from the south is
twice as large as the others. Reasoning from the
find of fragments of a stele showing a coronation
found on the podium, Goell reconstructed five so-
called stephanophoros stelae. She interpreted this
coronation scene as the transmission of power by
Mithradates I to Antiochos; Sanders suggests a
similar ceremony, but with Antiochos and
Mithradates II as protagonists.19 The rock behind
the platform with the stelae has been left
unworked, maybe because it remained invisible
behind the large orthostates.
At the north side of the podium a flight of nine
steps in tuffit has been constructed by Goell,
based on a protrusion in the lower line of steps.
It is situated in line with a stepped structure hewn
out in the rock at the top, in front of the base for
the right eagle and lion. The blocks may belong to
the podium structure; it can no longer be checked. Fig. 14. Nemrud Dağ, East Terrace, ancestor stele
At the south one step is constructed in the same (2001, photo E.M. Moormann).
way; but as far as we can see now, there is no
flight of stairs at this side.20 stelae standing here depicted the Persian and
In front of the podium Goell reconstructed a Kommagenean ancestors from whom Antiochos
so-called ‘sacrificial block altar’ parts of which are claimed to descend (fig. 14). Behind this row of
still in situ. The supposed altar is made from tuffit stelae, which must have given the impression of
blocks and slabs, material that lays scattered all a sculptured wall, there is a plinth with three ste-
over the area. Sanders already notes that this lae bases. The southern series contains 17 altars,
structure might not be original. The find of ashes whereas 13 bases for stelae have been preserved.
and other material in the structure suggests that The arrangement and measurements are equal to
local shepherds built the structure for own use those of the opposite side. Here the 17 claimed
from blocks and slabs lying around.21 Greek and Seleucidian ancestors of Antiochos
were depicted. Behind this row, at the south-west
3.1.4 Ancestor stelae and altars corner of the terrace, three bases and corre-
sponding slabs in tuffit are badly preserved.
As said, bases for stelae and corresponding altars
in front of them line the north and south borders 3.2 The West Terrace
of the terrace. The material used is limestone; the
northern row contains tuffit blocks in the rear side 3.2.1 Lay-out and architectural elements
of the basis.22 The north series of altars starts from
the mountain and counts 14 bases and 14 altars. The West Terrace is a flat space partly cut out of
The bases count two rows of blocks, always 14 the mountain, partly natural and partly con-
pieces. Every altar is composed of two tiers: at the structed artificially. At the eastern side there are
bottom two rectangular blocks, at the top a square the tumulus and, like on the East Terrace, in total
one. The altars measure approximately 80-85 nine colossal statues. Five large stelae, the so-
(height) x 70-75 x 70-75 cm. The (originally) 15 called Lion horoscope and four dexiosis stelae

83
 Fig. 15. Nemrud Dağ, plan of the West Terrace (© INF).

Fig. 16. Nemrud Dağ, West Terrace, overview from South (2001, photo E.M. Moormann).

84
Fig. 17. Nemrud Dağ, West Terrace, statue of Fig. 18. Nemrud Dağ, West Terrace, statue of
Kommagene, head standing on the terrace (2001, Antiochos, head standing on the terrace (2001, photo
photo E.M. Moormann). E.M. Moormann).

showing the king shaking hands with each of the statues whereas it must have been an empty space.
four gods, are standing beside the row of statues Goell mentioned an altar/offering table in front
to the north (see pp. 98-99). These five stelae were of the statues; this is, however, not visible.The
flanked at both sides by a lion/eagle pair as well. structure that can nowadays be seen in front of the
The south and (contrary to the East Terrace) the stelae (fig. 20, left) seems not original; we can not
west sides are occupied by bases for tuffit stelae distinguish five altars that were suggested by
with altars in front of them. As on the East Terrace, Sanders.24 There are no traces of paving. The liv-
the south side originally contained an orthostate ing rock rises considerably in some spots.
wall of 15 stelae showing the ‘Eastern’ ancestors; A path runs behind the stelae and statues
in this case the western side showed the 17 along the east side, similar to that on the East
‘Western’ ancestors (figs. 22-23). The northern side Terrace. To the west, the circular path around the
provides an access to the northern and the east- tumulus is hewn out in the rock.
ern terraces. Here Goell found a plinth with three The main entrance to the terrace was provided
socles and two altars in front. She also reports the by the hodos coming from the direction of Arsameia;
existence of relief fragments.23 She thus recon- fragments of an entrance stele had been preserved,
structed three or five stephanophoros stelae as on as well as its socle (see section 4.3.2).
the East Terrace. Also at this north side, at the
beginning of the slope, there was a large statue of 3.2.2 The colossal statues
a lion standing on a platform. Part of this guardian
animal and his base are still in situ. The open area The disposition of the gigantic statues is identical
between the base of the statues and the ancestor to the East Terrace, which is of great help in recon-
altars is now full of blocks fallen down from the structing them.25 There are a few differences in the

85
Fig. 19. Nemrud Dağ, West Terrace, detail: the base of the colossi covering the base of tuffit slabs of the stelae
and sculptures (2001, photo E.M. Moormann).

Fig. 20. Nemrud Dağ, West Terrace, the stelae and sculptures next to the colossi
(2001, photo E.M. Moormann).

86
details of clothing and rendering of the faces. The
pathos of the faces is enhanced in the Zeus and
Herakles who have a frown in their forehead and
bulging flesh over the nose; this feature being
shallower on the East Terrace. This and the beards
make them look older and more impressive than
the youthful Apollo and Antiochos with their
smooth roundish cheeks. The back of Kommagene’s
head is veiled as on the East Terrace. She has
remarkably protruding lips, her ears are set (too)
high and bear long jewels. The male figures wear
the Persian tiara, Antiochos has a feathered one
in Armenian fashion.26 This headgear consists of
a high, almost trapezoidal cap with three long
flaps covering the ears and the back. Antiochos’
ear-flaps are turned up and arranged over the Fig. 21. Nemrud Dağ, West Terrace, dexiosis stele
forehead so that they cover oneanother.27 His showing the king and Apollo-Mithras (2001, photo
well-shaped ears are visible. Along the front and E.M. Moormann).
the back of the tiaras runs a series of round discs
and all are embellished by a diadem tied together Nowadays the most northern stele lacks nearly
on the back, its fillets hanging downwards in the all its relief: on the right the outline of the kalathos
nape. These bands are studded with winged and the cornucopia of Kommagene is barely rec-
thunderbolts (Antiochos and Zeus) or round and ognizable. Fragments of the figures were found
lozenge discs (other males). Antiochos and Zeus by Humann and Puchstein and carried to Berlin,
have adorned neckbands, the former studded whereas Goell found the king’s head.33 The king
with winged thunderbolts, the latter consisting of had his attire adorned with lions and fruits, espe-
a torque ending into two lion heads under the cially pomegranates.
chin (figs. 17-18).28 The second dexiosis stele has the king and
Apollo-Mithras who wears a starred Phrygian
3.2.3 The stelae and sculptures in tuffit cap around which a sunburst with sixteen points
is shining. The king’s tiara has five feathers on top
As stated, a series of sculptures is placed to the and has a rich adornment of a lion walking to the
north of the colossi. Their base consists of one tier right in a frame of olive wreath at the lower side
of limestone and living rock topped by re-used and two lions in a lower band. The ear-flap,
tuffit slabs with erased inscriptions on the sur- adorned with olive leaves, is turned up and to the
face.29 At the southern side one can see how the front, the end of the other one is visible under-
second layer of the base for the colossal figures neath. Similar olive leaves are stitched on the
covers the tuffit slab (figs. 19-20). The stelae are king’s belt. His face has a round cheek, especially
inserted into this base by means of sockets. if compared with that of Apollo (fig. 21).
Whereas the northern pair of lion and eagle are On the third and largest slab Zeus is enthroned
now set apart, the other pair occupies its original on a mighty seat with animal-shaped front legs.
position.30 The dimensions of the stelae differ con- The heads of the legs are those of a lion, but with
siderably and the set looks rather irregular. pointed ears and horns.34 On top of the back a
The four northern stelae show a king in Persian pair of Greek eagles is seated, with spread wings
attire on the left shaking hands with a deity on and heads turned to their master. The tiara of Zeus
the right in the dexiosis ritual.31 It is important to is adorned with stars and, along the outline, beads.
see how the king, probably always the same per- At the lower edge runs a series of thunderbolts.
son, adopts iconographic elements of the person The king’s tiara has five feathers on top and the
in front of him in his own dress and attire - as will upper part has a thunderbolt and the foliate
become clear from the description given below.32 motive of oak leaves and glands. Underneath there
Zeus is much bigger than the king but, because is a diadem with winged thunderbolts. The belt
he is sitting their heads are at the same level and has similar oak leaves and glands, the boots have
heads, hands and other elements have similar thunderbolts.
sizes. Herakles is a little larger than the king and On the fourth stele Herakles holds his leonte
Apollo is slightly smaller. and his club near his left flank. Unfortunately

87
Fig. 22. Nemrud Dağ, West Terrace, stelae bases and Fig. 23. Nemrud Dağ, West Terrace, ancestor stele
altars, western side (2001, photo E.M. Moormann). (2001, photo E.M. Moormann).

details of the king’s dress are worn off, but he 3.2.4 Ancestor stelae and altars
must have had a lion and wine leaves as adorn-
ment on his dress.35 The west row of ancestor stelae and altars nowa-
The stele with the Lion horoscope is the most days still counts 12 bases for stelae and 12 corre-
famous piece. The animal strides to the right but sponding altars made from greyish limestone. The
turns its head with open muzzle and jutting blocks that had to contain the stelae rest on a base
tongue to the onlooker. A moon crescent covers of large blocks. They can be single or composed
his chest. Eight-pointed stars are scattered over of two halves. Behind them, lying on the rock,
the surface, three stars with sixteen points (in fact there are the bases containing sockets for the inser-
the planets Jupiter, Mercury and Mars) are shin- tion of the tuffit stelae. Some of them are hewn
ing over the back of the animal. They have their from the living rock. The altars are composed of
names written along the upper edge. We turn to two rectangular limestone blocks at the bottom
this stele more in detail in section 4.2.2 and a square one in the same material on the top.
These stelae look more elaborate and detailed They measure more or less 80-85 (height) x 75 x 75
than most of the slabs of the family members, if cm. At the northwest side there are four big tuffit
we assume that state of preservation plays no blocks, probably not in original position (fig. 22).
role. They are made with great skill and knowl- At the southern side of the terrace a similar dis-
edge of iconographical details; see for instance the position has been realised as on the western side.
adornments of the various dresses of the kings Ten bases are still extant, as are 11 of the corre-
corresponding with the gods. sponding altars. As stated above Goell calculated

88
 lar blocks from tuffit than from limestone, as can
be seen on the North Terrace.
As a matter of fact we are no longer able to
study the slabs: they are nearly all lost. The
detailed description by Young remains funda-
mental (fig. 23).37

3.3 The North Terrace

At the northern side of the passage between the

East Terrace and the West Terrace a series of 42
unworked stelae lies on the ground. They corre-
spond with 57 sockets; there are no altars here.
Together, the sockets form a continuous plinth in
which two pass-throughs have been made.
Puchstein and Humann already observed that the
slabs lack reliefs and texts. According to them the
stelae had never been erected and they inter-
preted the overall structure as a kind of fence
against the wind. Goell mentioned the find of
morter in one of the socket’s holes and is of the
opinion that the stelae stood upright. In her view
the ‘wall’ was finished (figs. 24-25).38
At the north side, near the East Terrace, a rec-
tangular structure in tuffit can be discerned. On
this platform stood a large guarding eagle of
which only fragments remain. Here, following
one of the pass-throughs in the long plinth, a road
leads down to the valley. Traces of it can clearly
be remarked on the basis of hewings in the rock.
Goell already observed a stele lying in the valley
below the North Terrace; indeed a logical place
Fig. 24. Nemrud Dağ, North Terrace (2001, photo for an entrance stele.39 Due to the fact that no
E.M. Moormann). remains of an inscription are visible on the stele
and given the circumstance that it could also
an original number of 17 sets. In one of the altars derive from the North Terrace, its function
the remains of a stele are still visible. The stelae’s remains uncertain.
bases are built up from limestone blocks managed
along the edge of the living rock, partly used 3.4 Building materials
instead of these blocks (after third block from the
east). They measure approximately 60-65 (height) Two sorts of material have been used for the con-
x 70-130 (width) x 85 cm (depth). The altars, made struction of the monument: grey to green slate-
of tuffit, are less regular than their southern coun- and sandstone-like stone, called tuffit by the geol-
terparts. Most of them are bigger and contain ogist Bernd Fitzner, and white-yellowish lime-
three layers of blocks and a cover plate on top. All stone. The first material, given its nature generally
in all, their measurements differ notably. The rather poor and badly preserved, served for mak-
sixth block from the east has a Greek inscription ing the stelae on all three terraces, some of the
in four lines on its eastern side.36 One may ask smaller freestanding figures and elements of the
whether the difference in format of the altars - architecture on the West and East Terraces. It was
contrasting to the regularity of all limestone altars probably dug in a valley some 1,700 m from the
on the East and West Terraces - had anything to East Terrace, near the place where recently a hotel
do with the importance of the person depicted on has been erected. Here quarry marks can be
the corresponding slab. However, such a division observed as well as ‘Rohlinge’.
cannot be noted on the East Terrace. Besides it The limestone is used for the colossal statues,
seems to have been more difficult to make regu- for most stelae bases and for the altars. It must

89
have come from the area itself, but is always 3.5.2 The colossal statues
thought to differ from that of the mountain itself
- very Karstic and splitting easily into plaques. Its On the West Terrace, the nine statues stand up to
origin is so far unknown.40 three of eight tiers; the other blocks tumbled down
The gravel of the tumulus stems from the in several moments, unknown to us. The East
mountain itself. The rock still standing on the Terrace has three of the five figures preserved up
West Terrace may have served as a quarry.41 to the shoulders (sixth row) and two more or less
up to the fifth tier, whilst the animals have fallen
3.5 Preservation, construction and technique off their platform except for the right-hand eagle
(three layers on base).
All architectonic and sculptural elements are in a The damages of the heads consist of holes in the
bad state of preservation, especially those hewn surface, which we do not encounter in the other
in the soft tuffit. parts of the statues. Therefore it seems that the
heads were hewn out of a different material. This
3.5.1 The stelae is not strange: this way they create an effect sim-
ilar to that of acroliths, where heads and other
Only some slabs are still erect (as a matter of fact nude parts are made of a different sort of marble
re-erected) and the few lying on the terraces hav- to create a contrast.
ing their original relief are but a small part of the The surfaces no longer show toolmarks apart
once existing number of slabs. Several sockets from some chisel marks in deep-lying places like
show remains of the stelae once erected in it and the corners between the wings and the breasts of
apparently broken off. the eagles in the fourth and fifth rows. In general,
As to the big stelae at the east side of the West the surfaces are worn by the exposure in the free
Terrace the surface is severely damaged: the air. The stone of the statues on the East Terrace is
Kommagene stele has poor remains only, the greyish, that of the heads on the Eastern Terrace
other show large lacunae in their reliefs. The cor- and of the statues on the West Terrace has a
responding series on the East Terrace is lost apart brighter colour and a smoother surface (where
from tiny fragments found by Goell and stored in not damaged). The material seems different from
the Adıyaman Museum. the limestone of the mountain on this place apart

Fig. 25. Nemrud Dağ, North Terrace (2001, photo E.M. Moormann).

90
from the lower elements of the Herakles and the
southern pair of animals on the West Terrace
which are hewn from the living rock.
The colossal statues are architectonic compos-
ites. They stand on a base cut out in the rock and
some even have the lower tiers hewn out of that
(West Terrace: Herakles, right-side animals). The
animals are composed of six tiers, the others in
seven (Herakles) or eight (others). The lower lay-
ers are like walls of a small building, the interior
space being filled with gravel and blocks of lime-
stone. All rows are composed like good masonry
by taking into account the joints between bound-
ers and stretchers. The front parts of the lower
layers consist of slabs modelled in relief. The
upper frontal layers of the figures, from the lap
onwards, are plastically rendered and the interior
sides have been hollowed out in order to make
them more manageable during construction. The
shoulders are entirely worked out from single
block, as are the heads. No concrete or mortar is
used to join the parts; their weight - especially
that of the top layers - was enough to keep the
blocks in position.
The heads on the East Terrace have protrusions
at the bottom and were set into sockets hewn out
in the shoulder blocks. The Antiochos head has a
pour channel at the bottom42 and the shoulder
piece of Apollo shows channels near the neck.
Zeus has a small channel on top of the tiara ele-
ment on which the tip had to be placed. The
tenon was either made out of the same block as
the head (Apollo) or produced separately and set
into cavities in both head and shoulder piece Fig. 26. Nemrud Dağ, East Terrace, the nomos on the
(Antiochos, Kommagene). As the tenon of the back of the colossi (2001, photo E.M. Moormann).
Apollo43 is some five cm smaller than the cavity,
we must assume that this head - and probably the
others - after being set into its position, were sta-
bilised by adding lead clamps in the same way
architects practiced in constructing buildings with
large blocks. The eagles’ and the lion’s heads and
all heads on the West Terrace have flat bottoms
corresponding in dimensions to the smooth sur-
faces on the shoulder pieces. The shoulder piece
of the Antiochos has a circle incised.44 Therefore,
these heads were simply put into position by lift-
ing. Their weight apparently sufficed and the cen-
tre of gravity was well calculated.
All blocks of the upper three layers have deep,
square holes, varying in number from two to four,
with sides of 5-7 cm and 9-12 cm deep (fig. 18).45
These are never positioned in the bottom or top,
but always in the front or the sides and they
served to hoist the blocks into their final position. Fig. 27. Nemrud Dağ, West Terrace, the nomos on the
The upper surfaces of most layers have shallow back of the colossi (2001, photo E.M. Moormann).

91
 Fig. 28. Nemrud Dağ, SIS level 2: plan of the West Terrace (© INF).

Fig. 29. Nemrud Dağ, SIS level 3: overview photograph of the West Terrace (2001, photo T.D. Stek).

92
indentations along the edges, probably for man- the East and West Terraces.46 Therefore, one of
aging the blocks in place. As to the heads one may this season’s main goals was the compilation of a
first think of the insertion of elements in metal, Site Information System. Its goals and use are
but more likely the holes served in all cases to lift explained below.
the pieces from the earth to their destination. As
the more obvious practise of leaving protruding 4.1.1 Goals
hoisting points in place, only to be finished of
after the final placement of blocks, was quite Drawing on the short description of the monu-
common in the Greek world; in the first instance ment, here presented in section 3, the SIS enables
we might be inclined to think of a re-use, or bet- the user to gain further knowledge about in the
ter re-placement, of the heads already in antiquity, monument him- or herself. It has different layers
after they had fallen of by, f.i., an earthquake. Be or levels, ranging from a general overview to a
that as it may, the construction of the statues was specific detail.
apparently planned and carried out by an archi- The most general level (1) is formed by the
tect, who employed the same methods as were topographical map (fig. 3). Many thousands of
used for the erection of large buildings. The large GPS points have been combined and show the
dimensions of the holes requires filling. In fact, tumulus with its remains and some of the char-
some holes in the columns at Karakuş have been acteristic features directly around it. Thus it
filled in with pieces of worked stone. It is strik- became clear that the top of Nemrud Dağ is not
ing, however, that building blocks on other 2150 m. heigh, as always stated, but 2206 m.47
Commagenian sites like Karakuş and Direk Kale From this overview the user can zoom in on
also show these holes. Do we have to reckon with one of the terraces. Level 2 provides detailed
the possibility that it was a Kommagenean prac- maps of the East- and West terraces with the posi-
tise to use lifting holes and leave them just as they tion of the statues, the fallen blocks and most
are? other elements. The level 2 maps consist of a com-
One may cautiously consider the possibility bination of GPS points from the topographical
that the surfaces of the figures had to be covered map (level 1) with handmade drawings. A docu-
with stucco and paint. The harsh climate, how- mentation of the actual situation on both terraces
ever, would make vulnerable this embellishment had not been done before (fig. 28).
and work would have to be redone every year. Level 3 takes the user into these maps by
overview photographs of the terraces. The ter-
3.5.3 The nomos inscription and other texts race’s maps can be visualised from different
angles while the position of the statues (rather
The backs of the human and divine colossi and evident) and the location of the different groups
the relief stelae are inscribed with Greek texts. of blocks (much more problematic) becomes
The colossi bear the nomos inscription, the relief clearer (fig. 29).
stelae mention Antiochos’ ancestors in a standard On the fourth level one can look at the colossal
formula. The lettering on both is carefully exe- statues themselves. Drawings show the statues’
cuted. There are no guiding lines. structures in different blocks and also indicate
The lay out of the large inscription differs which blocks are still in situ.
slightly per terrace and some wordings are dif- The statues’ tumbled down blocks can be stud-
ferent as well, but the columns are always the ied on the fifth level. It provides photos of groups
same. They present a sort of sub-chapter; para- of blocks and, after that, of individual blocks. It is
graphs are indicated by a long hiatus. The texts combined with a database that shows, i.a., our
on the back of the dexiosis stelae are less well exe- assignment of the blocks (fig. 30).
cuted (figs. 26-27, 34). Most of the individual blocks can also be found
on level 6, in which there are scale drawings of
4 THE AMSTERDAM UNIVERSITY RESEARCH blocks with sculpted details, from different per-
spectives, thus giving more detailed and precise
4.1 The Site Information System (with Tesse D. Stek information than the photographs.
& Ellen Thiermann) The 3Dimensional reconstruction combines
these different levels and will, in general, be of
As around 90 % of the blocks with which the stat- great help in obtaining a general understanding
ues have been constructed are still at hand, it of the monument. For reasons of convenience we
seems appropriate to reconstruct the colossi on thus labelled it level 0.

93
Level 0: 3D reconstruction drawn more easily than those on the West
Level 1: topographical map Terrace; besides the structure of the statues can be
Level 2: terrace plans better understood on the East Terrace because of
Level 3: terrace photographs their better preservation (see further section 4.1.3).
Level 4: colossal statues’ drawings The West Terrace is less well preserved and
Level 5: blocks photographs thus preliminary asked for a different approach.
Level 6: blocks drawings As a general overview of the West Terrace could
not be obtained by the laser scans (level 0 and 1
4.1.2 How to use the SIS are thus relatively incomplete here), it was
decided to compile a plan of the West Terrace
The general structure of the SIS as explained, will based on the (few) GPS points. On these fixed
be illustrated in this paragraph by a short descrip- points an imaginary grid was laid down; the
tion of the data now available for the East Terrace blocks were measured and drawn in by hand
and West Terrace. from this grid. The West Terrace plan presented
The East Terrace has been relatively well pre- here (level 2) is thus not 100% exact, but this
served. The gigantic statues are standing upright hardly matters because the fallen blocks are not
for the most part. The eagle and lion are, at both in situ. In order to get insight into the present
sides, largely collapsed and are less well pre- position of parts of statues fallen down it proved
served. Level 0 and 1 provide a general im- to work out very well. For the numbering of the
pression of the East Terrace as it is nowadays. On blocks see below. Level 3, overview photographs
the plan (level 2) the position of the blocks and of the West Terrace, makes the plan understand-
all other elements are precise, however not all able. The situation on the West Terrace being
(fragments of) blocks are visible. For the num- more complicated than on the East Terrace, the
bering of the blocks see below.48 Level 3 are pho- SIS user finds relatively numerous photographs
tographs providing an overview of the terrace on of it. Blocks are numbered from 1 to 152, again at
which nearly all blocks and other elements can be random but roughly from north to south. As on
seen. Blocks are numbered from 1 to 56 (for this the East Terrace, the database (Blocklist West
numbering see paragraph 4.1.3). Small or uniden- Terrace) gives our interpretation of the 152 blocks.
tifiable fragments are not always included. The For the statue/layer/block indications, and level
database (Blocklist East Terrace) gives our inter- 4 in general, the East Terrace statues and their
pretation of the 56 blocks, i.e. where we think they structure necessarely served as a parallel. The
were originally. The statue/layer/block indica- same indications are used. In level 5, photographs
tions refer to the (reconstructed) statues and their of the blocks, emphasis was mainly laid on
structure as presented on level 4. The statues are groups of blocks because many times individual
named from south to north: A (lion), B (eagle), C blocks do not show many, if any, specific charac-
(Antiochos), D (Kommagene), E (Zeus), F (Apollo), teristics. Hence, there are for now still many ques-
G (Herakles), H (eagle), I (lion). The base of the stat- tion marks in the database (Blocklist West
ues is always layer 1; layers are numbered from Terrace). For the same reason, and as explained
bottom to top; most statues have 7 layers. If lay- above, drawings of individual West Terrace
ers consist of different blocks, these have been blocks (level 6) are few. In general, blocks with a
given a letter. When facing a statue, the block in dimension of less than 0.2 x 0.2 m have only been
front, at the left side is designated a; counter included when showing sculpted elements.
clock-wise follow blocks b, c, d and e. The photos In the two blocklists thus seven columns are
of groups of blocks and individual blocks pre- shown. The blocknumber refers to the number of
sented in level 5 are also to be consulted together the block on the plan of the terrace. Then follow
with the database (Blocklist East Terrace). This the statue/layer/block indications as explained
way the user can study specific parts of the East above. Added are the name of the statue and of its
Terrace, also being able to check our interpreta- part in question; both just for visualisation as this
tion of the blocks. Level 6, drawings of individ- information is already present in the statue/
ual blocks, will be of great help in understanding layer/block indication. The last column indicates
the features characterising specific blocks. We the state of preservation, viz. a fragment or a
decided in this stage to provide a relatively large whole block. The blocklists are connected with
number of drawings of the East Terrace blocks (in another part of the database, i.e. the photo/draw-
comparison to the West Terrace, see below) for inglist. This contains the following information:
practical reasons. The East Terrace blocks can be the code of the photo with the corresponding

94
 Fig. 30. Nemrud Dağ, SIS level 5: block photos of the East Terrace (2001, photo T.D. Stek).

blocknumber; the view (side, detail, overview, In the first place the documentation of the sta-
etc.); a table indicating if there is a drawing of the tus quo of these fragments is necessary before any
block; and remarks. With the photo code the first restoration, conservation or reconstruction work
two letters logically refer to the East Terrace (E) was to be undertaken. Secondly, detailed mea-
or West Terrace (W), the following figures indi- surements are needed for our reconstruction pro-
cate the blocknummer (in case of overviews the posals, as they can show which blocks fit in and
central block has been taken), the final letters which do not. Details documented by these draw-
indicate the view (ba = back, fr = front, si = side, ings can also provide further insight into tech-
ls = left side, rs = right side, bo = bottom, to = top, niques of sculpturing and engineering used; also
gr = group photo, xx = unknown). General the lifting holes, for instance, are measured and
overviews, blocks of the statues’ bases and the indicated. And finally, all earlier publications on
like may have alternative codes; these are how- Nemrud Dağ lack detailed drawings of the tum-
ever always evident (East Terrace overview 1, bled down blocks.
West Terrace base F, etc.). The blocks were given individual numbers
which are independent from their possible iden-
4.1.3 Block drawings tification, but generally in correspondence with
their present position in the field: from left to
As explained above, level 6 of the SIS consists of right when facing the statues.
handmade drawings of the blocks of the colossal Drawings were made from the flat bottom- and
statues on the East and West Terrace. For several top surfaces of the blocks, as these form the junc-
reasons we aim to document the fallen blocks in tions with the other tiers of the colossi, and for
detail and as completely as possible. that reason are indicative of their position in the

95
statue. Further drawings document the sculp- study of the literature available up to now. Some
tured side(s) of the blocks. Unfortunately, due to of the questions are dealt with in the following
the position of many of the blocks in the field, it sections.
was not always possible to provide all three (or
more) views. If possible, we reconstructed the 4.2.1 Cosmological orientation?
outline of damaged or invisible parts of blocks in
dotted lines. All drawings were made on a 1:10 The monument shows a strong irregularity in the
scale. orientation of its elements. No line exactly runs
High resolution digital photographs were N-S or E-W. As to the East Terrace one may
taken from the same angles as the drawings to observe that Regulus, one of the fixed stars, rises
add information about texture and colour (level at square angles with the set of colossal statues;
5). As said above, block drawings and photo- their display thus corresponds with Regulus.
graphs are linked in a database (Blocklist West Therefore, one may assume that the cosmos
and East Terrace), where our proposal for the played a great role in the planning of the various
identification is given as well. All drawn blocks elements, perhaps also including the stelae and
are indicated in red on the drawings of the colossi their altars.49
(level 4). During the 2001 campaign the drawings In general, however, how attractive the possi-
of the blocks lying on the East Terrace were com- bility of a cosmological orientation may seem, cer-
pleted. The majority of the blocks on the West tainly in view of the cosmos’ large role in
Terrace are to be drawn in successive campaigns. Kommagenian culture and perception in general,
we have to remain reticent with these kinds of
4.2 Some new observations interpretations. As can be concluded from our
short description of the monument, the natural
In compiling the SIS, epigraphic and archaeolog- environment, with its possibilities and limitations,
ical interpretative research was carried out. The played a large role in the monument’s building.
archaeological questions partly raised from the Always, and perhaps primarily, this kind of prac-
demands mentioned, partly from the preliminary tical reasons should be taken into consideration.

Fig. 31. Nemrud Dağ, West Terrace, the Lion horoscope (2001, photo E.M. Moormann).

96
4.2.2 The Lion horoscope: proposal for a new dating meinen Zügen den Zustand des Sternenhimmels
(Maurice Crijns) zur Zeit des Horoskops wiederzugeben; denn in
einer vollkommenen Darstellung desselben müßte
One of the most spectacular finds Humann and sowohl der Teil des Zodiakalzeichens, der bei einer
Puchstein made on the West Terrace was a large zeitlich bestimmten Beobachtung über dem
slab representing a huge lion, speckled with 19 Horizont emporsteigt, ausdrücklich bezeichnet als
stars and with a moon crescent on his chest (fig. auch der gleichzeitige Ort aller Planeten nach
31).50 Each star has eight points. Apart from a Graden oder sonstwie genauer angegeben sein.’53
small difference, the positions of the 19 stars rep- As Lehmann’s constellation could never have
resent the constellation of Lion as described already been visible in reality, his calculation was gener-
in antiquity in the Ephemeris of Eratosthenes.51 ally rejected. In his fundamental work on horo-
Three larger ‘stars’ with 16 points are depicted scopes Otto Neugebauer calculated it as the con-
at the upper left edge. They represent the planets stellation on July 7 of the year 62 BC; a date that
Mars, Mercury and Jupiter, as becomes clear from could possibly be connected with the coronation
the inscriptions above them. of Antiochos.54 Neugebauer reconstructed the
The two scholars immediately understood that planetary positions as follows:
the depiction on the slab represented a horoscope.
The stars have a fixed position in the sky. Each Planet Longitude Latitude
evening one can see them at the same place, while Saturn 43 –2.1
the Sun, the Moon and the planets are moving Mars 128.7 +1.1
day by day with different velocities. The stele is Mercury 128.5 –0.8
the frozen picture of the positions of the heavenly Jupiter 114.9 +0.7
bodies at a certain moment at a certain date. Moon 132 +4.5
Regulus 120.6 +0.35
Scholarly opinions: Lehmann and Neugebauer Sun 102 0
Venus 85 +0.1
Humann and Puchstein consulted the astronomer
Paul Lehmann to make a calculation of the date Virgo |146 Lion 106 | Cancer | Gemini
the horoscope would represent.52 Lehmann was
asked by Karl Humann to take into account the first
half of the first century BC only, since Humann
assumed that the Lion stele referred to Antiochos.
+5
+4
+3
" Moon

Lehmann found several dates of which the 17th of +2 Mars Jupiter

July 98 BC. seemed most likely. Puchstein thought +1
R

this date to be the birthday of Antiochos, but, as 0 | | |  | | | | | |
the inscription states that the 16th of Audnaios (a 150 140 130 120 110 100 90 80 70 60
date in December or January) was the date of –1

birth, he concluded that the calculated date was –2 Mercury
the day of his conception. Hence, Antiochos –3
should have been born somewhere at the begin- –4
ning of 97 BC as a seven months child. –5
This theory has three major problems:
1. the planets Mars, Mercury and Jupiter are The great advantage of this theory in comparison
standing before Regulus and not, as depicted with the proposal by Lehmann is that all planets
on the Lion stele, after Regulus; including Mercury could be seen during the
2. the Moon is not in conjunction with Regulus night. However, the reconstruction of this con-
but has passed Regulus four days before; stellation poses several major problems when we
3. the Sun is in conjunction with the planets. That look at the picture on the Lion stele:
means that with Sunset, Mars, Mercury and 1. the Moon has passed Regulus the day before;
Jupiter are descending. So these planets cannot 2. the Moon has a positive latitude, thus passes
be seen at all. Regulus from above and not from below;
Perhaps as a justification for the imperfect con- 3. the Moon has passed all the planets;
stellation found and the speculative conclusions 4. Mars is in conjunction with Mercury, whereas
attached, Puchstein wrote: ‘Ein derartiges the stele shows a sequence of Mars-Mercury-
Reliefbild vermag allerdings nur in sehr allge- Jupiter;

97
5. Jupiter is in front of Regulus and will need from its geocentric position, owing to the
another 24 days to pass Regulus. observer’s location off the line from the centre of
Goell associated the date found by Neugebauer the Earth to the body. This effect is called paral-
with the foundation of the hierothesion on lax and occurs especially in case of the Moon.
Nemrud Dağ.55 Wolfgang Haase saw the stele as The following picture shows the sky seen from
the expression of a personal katasterismos of the the West on the evening of the 14th of July 109 BC:
king.56
Virgo |146 Lion 106 | Cancer | Gemini
A new calculation
+5
We can conclude that from all hypothetical con- +4
stellations between 100 and 50 BC, this one equals +3
most that of the Lion horoscope, but that it is not +2 Mars Jupiter
identical and thus remains unsatisfactory. There-


+1
fore, a reinvestigation has been executed, in which 0 |Mercury
| R | | | | | |
the assumption that the constellation has to be 150 140 130 120 110 100 90 80 70 60
searched in the era between 100 and 50 BC has –1
been abandoned. Starting from 1 AD backwards a –2
search was executed, making use of the Ephemeris –3
of Tuckerman.57
Jupiter needs about 12 years to pass the eclip-
tica (i.e. the, in fact imaginary, circle in the sky the
–4
–5
Moon
"
sun seems to pass through within a year), Mars 2 As to the visibility of the planets the distance to
years and Mercury 1 year. The Moon passes the the Sun is of importance as well as the brightness
ecliptica in 1 month. First, those periods were (magnitude). Jean Meeus was so kind to calculate
selected in which Jupiter was in the constellation following magnitudes:
of Lion as depicted on the stele, according to
Humann and Puchstein ranging from 106 to 146 Planet Mars Mercury Jupiter Regulus
degrees of the ecliptica. From these periods those Magnitude +2.0 +0.2 -1.3 +1.3
were singled out in which Mars was also moving
through this part of the ecliptica. And the same - Mars could be seen with the naked eye.
had to be true for Mercury. The second selection - Mercury cannot be seen with the naked eye
criterion was the sequence of these planets most of the time. However, on this particular
according to the Lion stele: Mars-Mercury-Jupiter. day Mercury had reached almost its maximum
As a result the constellation of the July 14, 109 BC distance from the Sun (being 28 degrees) and
(Julian calendar) was found.58 was therefore well visible with a magnitude of
The following planetary positions are derived + 0.2.
from the Tuckerman Ephemeris by a linear inter- - Jupiter with a magnitude of -1.3 had the great-
polation for 19h00 Babylonian Time, being 18h34 est brightness and could be seen as a big dot in
Local Time at Nemrud Dağ (co-ordinates 37°59’ the sky.
N and 38°45’ E). - Venus had set before the Sun and could not
been seen.
Planet Longitude Latitude - Saturn could be seen very well. However, he
Saturn 184.06 +2.31 stood far away from the other planets, close to
Mars 135.24 +1.09 the south at its culmination.
Mercury 133.21 –0.6
Jupiter 126.38 +0.93 The welcoming of the gods
Moon 119.82 –4.52
Regulus 120.6 +0.35 To summarise our hypothesis: the weeks before
Sun 108.38 0 the planets had passed one after another Regulus:
Venus 102.49 –6.79 Jupiter on June 10, Mars on June 21, Mercury on
the July 6 and finally the Moon on the evening of
The Moon is not corrected for parallax. To an July 14.59 Interestingly, the planets Jupiter, Mars
observer on the surface of the earth, the position and Mercury (the gods) passed Regulus (the king)
of a body at a finite distance will generally differ from above, whereas the Moon (the land of

98
Kommagene) passed from below. For the people Regulus was to be seen in the evening sky, but it
of Kommagene it must have looked as if the gods is not sure that the conjunction of Moon and
Zeus/Oromasdes (Jupiter), Artagnes/Herakles/ Regulus was visible, although 24 hours had
Ares (Mars), Apollo/Mithras/Helios/Hermes passed after New Moon. The time lapse between
(Mercury) and Kommagene (Moon), had come Sunset and the descent of the Moon was about 17
one after another to congratulate Regulus, the minutes. If the Moon-Regulus conjunction was
King. visible from the top of the mountain, it was only
The welcoming of the gods is (also) depicted for a very short time: just before the descent of the
on the four dexiosis stelae at the left of the Lion Moon at 19h37 Local Real Time. In the minutes
stele (see paragraph 3.2.3). Furthermore the 10th before, the complete constellation as depicted on
of Loos (i.e. July 14) is mentioned in the inscrip- the Lion horoscope, was in place. Next, the con-
tion as the date of the coronation of the king. stellation disappeared from the sky part by part;
The exact time of the constellation was Sunset, Regulus descended at 20h04, Jupiter at 20h25,
19h20 Local Real Time (=16h45 UT). To calculate Mercury at 20h43 and Mars at 20h53.
this time in greater detail, calculations are This particular phenomenon makes it possible
required by linear interpolation based on the to determine not only the date but also the time
Tuckerman Ephemeris. We could make use of the of the constellation, viz. 109 BC, July 14th, at
astronomical software developed by Jan Rademaker approximately 19h35 Local Real Time.
to do this. The planets’ positions can be recon-
structed as follows: The originator of the Lion horoscope

Planet Longitude Latitude It has been assumed that Antiochos erected the
Mars 135.24 +1.08 Lion stele in the first half of the 1st century BC.
Mercury 133.33 +0.06 Humann and Puchstein expressed their doubts
Jupiter 126.37 +0.93 about the dating of the stele because of the clumsy
Moon1 120.13 –5.72 style: ‘An einem etwa um das Jahr 40 v. Chr. ent-
Regulus 120.60 +0.35 standenen Werk von hellenistischer Hand würde
Sun 108.40 0 man Eigenschaften wie die eben geschilderten
gewiß nicht erwarten.’60 A dating to 109 B.C. per-
Seen from Mount Nemrud the positions are as haps could explain why the stylistic features dif-
follows: fer so much. Some scholars proposed that there
has been an earlier sanctuary on top of Nemrud
Planet Azimuth2 Height Dağ constructed by Mithradates I Kallinikos, the
Mars 98.93 +17.33 father of Antiochos. The dating of the Lion horo-
Mercury 99.87 +15.37 scope provides us a new clue to corroborate this
Jupiter 106.05 +11.75 hypothesis (see also section 4.2.3).
Moon 106.70 +2.65
Regulus 110.22 +7.63 4.2.3 Chronology: a new dating frame?
Sun 119.37 0.453
In all extant publications on Nemrud Dağ the
1 The position of the Moon has been corrected monument is considered a work ordered by one
for parallax. single man, commemorating the glory of himself
2 Azimuth: South 0 degrees, West 90 degrees, etc. and, in the second place, that of his ancestors:
Height: 0 degrees = horizon. Antiochos I. This king reigned over the small state
3 The visible Sunset and Sunrise is influenced by of Kommagene from 69 to 36 BC61 and, as stated
the bending of the light beams by the atmos- in his inscription, he erected the statues of Zeus,
phere. This phenomenon is called atmospheric Kommagene, Apollo, Herakles and himself when
defraction. The Sun is lifted approximately + he was old of age.62 His father makes part of the
0.5 degree at sea level and approximately 0.4 ‘Ahnengalerie’ and his son Mithradates II is men-
degree at an altitude of 2,000 m. So, the visible tioned as one of the young persons on the slabs.
Sunset seen by an observer at Nemrud Dağ Some scholars suggest that the monument was
occurs when the Sun is approximately 0.4 never finished.63 The Antiochos’ head on the East
degree below the horizon. Sunset is therefore Terrace with not worked-out ears and the absence
19h20 Local Real Time with a possible devia- of any find of ceramics and other mobilia were
tion of maximum 2 minutes. the main arguments.

99
 The evidence of the Lion horoscope and the few tuffit bases at the southern side of the East
(possible) presence of a second horoscope in the Terrace. All tuffit elements, including guardians
shape of the colossi (see 4.2.7) raise the question in the shape of threedimensional lions and eagles,
whether we have to reckon with the possibility of were erected on Nemrud Dağ in this period. This
two different monuments. Humann and Puchstein work must have been carried out at an indeter-
already observed the difference in style between minable moment during the reign of Mithradates
the stelae and the statues. A remarkable feature in I, viz. 109-69 BC. The series of stelae on the North
this respect is the use of two different materials: Terrace were erected, but not worked with texts
tuffit for the stelae and part of the altars and or reliefs.
building elements and limestone for the colossal ‘Limestone phase’: Antiochos enlarged the
statues and greater part of the altars and the monument, turning it into his tomb. He con-
slabs’ bases. structed the artificial tumulus, hewed out a ter-
Because of difference of material and differences race over the old one on the East Terrace and
in form and style of the stelae in respect to the enlarged the space on the West Terrace by cutting
limestone statues one may suspect that the tuffit out the stone until the part still rising at the south
elements form an independent ensemble, not side. The addition of the gigantic statues on the
made in combination with the statues. If they are West Terrace created a new situation for the old
considered separate from the statues they seem stelae. These had to be re-placed into new bases,
older and may belong to the reign of Mithradates located in a position fitting to the new setting and
I: the (new) date of the horoscope is the main clue therefore the row along the west side was created.
for this conclusion. Besides, the inscriptions of The series on the south side probably held their
Antiochos on the back-sides of the dexiosis stelae original position and kept the old altars in tuffit.
on the West Terrace are written over erased older The rows on the East Terrace were restored: old
texts, which have left an irregular striated, not blocks are visible under the northern row (fig. 19).65
smooth surface (fig. 34).64 It cannot be established Antiochos erased the old text of Mithradates I on
whether the date on the Lion horoscope is also the dexiosis stelae - it could have been a nomos
the date of production. according to Waldmann and Goell - and wrote his
In view of the new date, one should consider own texts on them in a rather clumsy style (espe-
the possibility that Mithradates had himself cially if compared to the nomos on the backs of the
erected a monument on the mountain in 109 BC, colossi). This gigantic work was not completed as
or soon after. Alternatively, Antiochos may have we can see in various parts of the colossal statues
transferred a monument that his father had (heads of Apollo and Antiochos on East Terrace,
already built elsewhere, to this barren spot. lack of holes for heads in neck pieces on West
Points of consideration are: Terrace). As the king states in the large text that
1. the two different constructing materials, tuffit he was old, we must assume that the work began
and limestone; in his late years, let’s say the last decade of his
2. the traces of erased inscriptions on the stelae reign (45-36 BC).66
and their bases next to the statues on the West ‘Aftermath’: Mithradates II ascended the throne
Terrace; after a struggle with his brother. He apparently
3. the date of 109 BC indicated in the Lion horo- had not been the first candidate. He left Nemrud
scope; Dağ as it was and started work on a new tomb at
4. construction details; Karakuş, where his mother Isias (maybe Isias
5. the statues are mentioned in Antiochos’ large Philostorgos), his sister Antiochis and her daugh-
text, but that the slabs are not. ter Aka would be buried.67 His respect for his
Considering this, we suggest that the monu- father had perhaps shrunk after the problems
ment was not planned and executed at one single about the succession. Sanders even suggests that
moment in time, but was executed in several Antiochos was murdered.68 Another reason may
stages. As a working hypothesis we propose the be the diminished prestige of the Kommagene
following chronological framework. kingdom; or a combination of either.69
‘Tuffit phase’: Mithradates I constructed a ter- Also other scholars working on the mountain
race on the East Terrace on which he erected a already hinted at a more extensive chronological
series of stelae, and built a base with stelae on the framework. Already Waldmann stated that there
West Terrace. The stelae of the ancestors and him- was perhaps an earlier sanctuary, built by
self as the last member of the family were placed Mithridates I, as did Şahin.70 Goell had an elabo-
in a display we cannot reconstruct apart from the rated hypothesis on an earlier sanctuary and

100
Jacobs and Hoepfner talk in a general vein about lion and eagle statues. We must conclude, then,
‘eine frühere Planungsphase’.71 that we should approach Goell’s conclusions with
It may be best to reserve judgment on this care. We can conclude that there may have been a
hypothesis at this stage of our research. In the horoscope stele on the East Terrace, and that it
first instance the presence of two different kinds stood with some undetermined stelae on the
of material may well be explained on practical podium structure. In view of the structural simi-
grounds alone: hard stone was used for the stat- larities between the East and West terraces, Goell’s
ues, whereas the softer limestone was easier to reconstruction is in itself not unreasonable, but it
work on for the reliefs. One also has to count with should be remembered that we cannot determine
the possibility that the Lion horoscope, represent- the exact nature and sequence of these stelae.
ing - as we have argued - a date in 109 BC, was On the basis of our hypothetical chronological
made later on to commemorate an important his- framework formulated above, one may cautiously
torical event. Lastly the interpretations by Hoepfner propose the following sequence of situations.
have to be noted in this respect. In the end this Mithradates I created the East Terrace with a
author considers Antiochos’ building activities as plateau cut out in the mountain. He erected slabs
a ‘Kunst-Mythos’ and notes that the German with representations of himself in the act of dex-
excavations in Arsameia have never succeeded in iosis with Zeus, Kommagene, Apollo and Herakles
finding the older phases that Antiochos claimed as well as the Lion horoscope and the coronation
in his texts: ‘Und wer unter den Reliefbildern stelae. In front of it the stepped structure was
ältere Stücke ausmachen möchte, entspricht zwar constructed. Its purpose cannot be determined:
den Erwartungen des Königs, befindet sich aber was it an altar, a podium or something else? (see
wahrscheinlich auf dem Holzweg.’72 The scholar further below, section 4.2.5). The other sides of the
looking for different phases in Antiochos’ build- terrace were flanked by the series of ancestors, the
ing program is therefore warned. 15 Persians at the north side, the 17 Greeks at the
south side.
4.2.4 The lower ensemble on the East Terrace Mithradates’ son Antiochos left the old situa-
tion more or less intact, but replaced the fragile
Goell found numerous mostly tiny remains of ste- bases of the stelae by limestone blocks. More
lae at the foot of the huge statues on a sort of important is his amplification of the ensemble: he
podium over a flight of what she thought to be cut out a second terrace over the niche built by
three steps. She reconstructed a row of stelae simi- his father and placed here the series of nine colos-
lar to that still standing on the West Terrace but sal statues. The old ceremonies may have contin-
could not establish their sequence. Other stelae may ued in the old form and be enriched by special
depict a coronation ceremony (see paragraph 3.1.3). elements regarding his own monument. It is not yet
This unit forms an ensemble on its own and entirely clear how the ancient visitors approached
nowadays has a low flight of steps in tuffit at the these new colossi. Their platform has three steps
right and left hand sides. Goell interpreted these in front and some more at the side of the northern
as steps leading to the higher terrace with the eagle. Here one could expect a prolongation of the
colossal statues. But from observations on site it flight of stairs built by Mithradates I as suggested
has become clear that Goell’s interpretation by Goell, but it seems that this would have been
should probably be rejected. These steps, irregu- simply too high for stairs of this type. Maybe,
lar though they are, belong to the construction of something like the modern ramp that was built at
the lower terrace.73 the north side, also existed in antiquity, which may
Goell’s reconstruction of five stelae standing on have provided a connection between the North
this structure, four of which with a dexiosis and and West Terraces. It is unclear whether there were
one with a Lion horoscope, the same as on the stairs at the southern side, the place where people
West Terrace, is generally accepted.74 It is impor- entered the terrace via the twisting staircase.
tant to realise, however, that this reconstruction is Antiochos embraced, as it were, his father’s
only based on the find of some tiny relief frag- monument by building over it his own series of
ments. Some of these indeed seem to be part of a statues. Their upper outline reflected at the same
horoscope; the other possibly belonged to dexiosis moment the contour of the tumulus built by him
stelae.75 Goell’s text seems to suggest that also over the mountain itself. Whereas Mithradates I
part of the stele of the lion was found; but she sought contact with the gods by means of dexio-
presents no evidence in support of this, nor does sis, Antiochos placed himself at the same level as
she produce material evidence for the flanking the gods.

101
4.2.5 The ‘Fire Altar’ on the East Terrace: true or a check impossible. The idea of an altar may well
false? have been inspired by a - not written - comparison
with the Zeus altar at Pergamon or other monu-
Humann and Puchstein found two parallel rows mental examples of this era.80
of steps running east-west at the eastern edge of
the East Terrace, five steps at the south and four 4.2.6 The West Terrace in its oldest shape
on the north, together with a wall running north-
south east of these steps and a ‘dromos’ directed In this case our hypothesis concerning the phases
towards the statues.76 This situation is also of construction, does not help us any further in
depicted on Hamdi Bey’s map. Goell, on the basis determining how the old situation differed from
of information of her workers, thought the ‘dro- the new. At one glance, it is clear how the tuffit
mos’ to be a modern construction, a sort of bait base of the stelae at the right side is covered by
for catching kekliks, a local sort of pheasant, and the base of the statues. The steps cut from the liv-
destroyed it.77 The other remains were recon- ing rock under the lower layer of the colossi reach
structed as three of four sides of a large stepped a slightly higher level than the steps under the
altar. As such the reconstruction of the stepped stelae (fig. 19).81 This implicates an alteration or
elements and the retaining wall into a square an adaptation to a new situation.
architectonical structure (fig. 4) is not unreason- The juxtaposition of the couples of lion and
able. The old maps suggest a straight outline at eagle in both tuffit and limestone can possibly be
the west side of the steps. Most recently, explained as a sign of philia, piety on the part of
Hoepfner did agree with this reconstruction and Antiochos towards his father: he wanted to keep
considered the altar as an essential part of the complete the old monument. In a way, even the
sanctuary.78 juxtaposition of the same gods in two forms
However, this reconstruction does not imme- (enthroned and in the act of dexiosis) is - to say the
diately allow an interpretation of the platform as least - odd.
an altar. The shape of the monument and the As said in the description (section 3.2.3; fig. 20),
notion of a stepped element at one side of the ter- the series seems to be irregular because of the dif-
race can be explained in various ways. One could, ferent measurements that were used, and because
for instance, imagine the king himself, with local of the carefully rounded outlines of the top that
dignitaries around him, sitting under a kind of rarely correspond to the adjacent slabs. The
velum, diametrically opposed to the figure of Zeus, Kommagene and Apollo stelae seem to form a
looking at ceremonies in the courtyard. couple; their upper outlines form a shallow curve.
We suppose to treat the structure destroyed by The other slabs have similar curved tops, but do
Goell as an original element for the following rea- not match. If we were to take into account the
sons: possibility that there used to be counterparts, we
1. the ‘bait’ has regular, smooth inner walls and would have to assume at least two other pairs of
rough outer ones according to the very accu- gods, one matching the Zeus stele, and one
rate map by Humann and Puchstein (Hamdi matching the Herakles stele. The Lion horoscope,
Bey only has straight lines); with its rounded left upper corner where the
2. the walls stand at square angles to the ‘altar’; inscription follows the outline, may have had its
3. the structure points exactly towards the colos- own counterpart as well, for instance the constel-
sal statue of Apollo and is not positioned in the lation Eagle (Scorpion/Aquarius).
middle of the ‘altar’ wall. If we are right in distinguishing a construction
in two stages, we are, however, unable at present
Utecht’s geophysical research evidenced the pres- to reconstruct the first phase. Antiochos appar-
ence of a cavity under the ‘altar’.79 One may, ently had to change the entire area when he
therefore, ask whether the parallel walls formed wanted to add the colossal figures on this narrow
a sort of dromos leading towards the hollow place. The unsatisfactory placement of a row of
rooms under the terrace. That does not mean that ancestor stelae along the western side can have
we suggest the presence of the tomb chamber been the only solution available.
here: the cavities seem to be too small, but one
may think of an offering place. Its nature can only 4.2.7 The sequence of the colossal statues
be investigated if Goell’s altar will be demolished.
In conclusion, we fear that Goell’s decision was It has been a matter of scholarly debate where
incorrect. The destruction of the evidence makes the Apollo and Antiochos heads have to be

102
located. Humann and Puchstein meant that Eagle = Scorpion/Aquarius
the king had his seat between Zeus and Kommagene = Moon
Herakles; they only knew the unbearded head Zeus = Jupiter
nowadays still standing in front of the terrace Apollo = Mercury
(fig. 11) and had not yet discovered the other Herakles = Mars
one lying at the backside of the colossi.82 This Eagle = Scorpion/Aquarius
hypothesis was discarded by Goell and since Lion = Lion
the important essay by John H. Young (1964) it The gods seem to look towards their own stars.
is generally assumed that Antiochos is the fig- Mercury is only visible when standing at its
ure sitting at the extreme left.83 Young’s main maximum distance from the Sun. The combina-
argument is the iconography of the Armenian tion of all these ‘stars’ visible at the same
tiara with the upturned flaps. He also points at moment is in 87 BC. This constellation may rep-
the order of the figures as mentioned in the resent the personal horoscope of Antiochos, sim-
inscription (East Terrace IIa, 9-17): apparently ilar to that on the Lion stele discussed above. If
the king started from the main figure, Zeus, the Lion horoscope was also actually made in
listed the figures to the right (Herakles is sure) 109 BC and thus has to be connected to
and concluded with those on the left. There are Mithridates I, this implicates that the younger
some additional arguments in favour of Young’s king wanted to imitate and surpass his father in
proposal: one and the same time.
1. the actual position of the heads of Apollo and The second hypothesis concerns the symbols of
Antiochos;84 the four elements:
2. Antiochos forms a parallel to the semi-god
Herakles in sharing the gods’ community and Zeus with his eagle = Air (eagle as symbol)
‘adds’ himself to them; Apollo-Mithras- = Fire (bull as symbol)
3. the usual position of a ktistes; Helios-Hermes
4. the height of the statues in relation to one Artagnes-Herakles- = Earth (lion as symbol)
another. Ares
Kommagene = Water (she is patris
The choice of these particular gods is not explic- pantrophos, which can
itly explained apart from that of Kommagene as only be true if there is
the symbol of his patris pantrophos. The choice for water).
Zeus-Oromasdes, the father and leader of the
gods, is obvious and Artagnes-Herakles-Ares In this way the set of four gods was complete and
apparently serves as a role model for Hellenistic Antiochos could share the four-some. The pairs
kings (see infra). The association to Ares is per- of animals embrace the ensemble and repeat the
haps due to an important victory like that at important link between Zeus and Herakles and,
Pharsalos in 48 BC.85 This does not seem very now, Zeus and Antiochos and/or Herakles and
likely, however, in view of the fact that nowhere Antiochos. Moreover, sitting next to Kommagene,
in the complex a direct hint at contemporary pol- Antiochos forms a couple with her as his own
itics can be found, as well as the circumstance that country. In this reasoning, Apollo remains the
dynastic presentation, and not self-representation least explicable figure despite his many other
seems to come to the fore most dominantly on names: was here the Persian element of Mithras
Nemrud Dağ. predominant?
There is no specific reason known for the inser- Finally, it is important to notice that there may
tion of Apollo-Mithras-Helios-Hermes. We pro- also be a more general explanation for the choice
pose two more hidden explanations, an astrolog- of these gods. In the Hellenistic period in general
ical and a symbolical one.86 especially the trio Zeus, Apollo and Herakles was
The first idea, originally put forward by M. related to kingship and the ruler, as is witnessed,
Crijns, brings us back to astrology and astronomy. amongst other things, by the prominent place
All figures seem to symbolise a star in the con- these gods hold in Alexandrian poetry. By choos-
stellation and the series may thus form a specific ing this trio Antiochos thus placed himself in a
situation like that in the Lion horoscope: royal, Hellenistic tradition, presenting himself as
a typical Hellenistic ruler. The portrait of Komma-
Statue = heavenly body gene, and that of the king himself, added a local
Lion = Lion touch.

103
4.2.8 Style and iconography of processional roads and markers, part of the
sanctuary and its conception. To understand what
The colossal statues on Nemrud Dag˘ are coarse was happening on the terraces, we therefore
works of sculpture apart from the heads that show thought it necessary also to study this direct,
a great interest for detail despite their huge mea- wider context; especially because some interesting
sures. Thrones, boots, and clothes: few details enrich questions remain to be answered.89
the figures. The only additions are the fruits in
Kommagene’s lap, her cornucopia, Herakles’ 4.3.1 Introduction
stubbed club and the mane of the lion. In sum, the
surfaces are plain for the most part. The East Terrace The area directly surrounding the mountain was
statues are still simpler than their counterparts. systematically investigated for the first time in the
The heads, as said, show a certain degree of 1950’s by Goell and members of her team, and
detail and some of them even are refined works mainly by the topographer Heinrich Brokamp. In
despite their measures. The head of Antiochos on mapping the mountain and monument, Brokamp
the East Terrace has remained unfinished as has also made a rather basic but reliable map of the
been observed before by Young.87 Antiochos has circle with a diameter of around 1000 m around
his ears not worked out and the lappets of his the tumulus. The main information provided by
tiara are left unadorned. He has exceedingly this research was the existence of two proces-
roundish cheeks and the small lips stick out of the sional roads and markers (inscribed and deco-
smooth surface like the lips of a fish. The mouth rated stelae) as well as a route to a spring in the
is closed, whereas all other figures have parted valley where the nearest source of water was be
lips.88 It seems plausible, therefore, that Antiochos to found. No traces at all were found of what we
was waiting for the last treatment, especially con- could call construction places and activivities (as,
cerning the ears. His face contours are roundish for instance, accomodation for the people build-
and should have been reduced slightly as well. ing the monument) or other facilitaties for a sanc-
The heads of the colossi of both terraces are tuary in use (as, for instance, lodging places or lit-
similar only, not identical. The beards of Zeus and tle stations on the way up).
Herakles of the West Terrace are richer and con- The aims of the 2001 survey around the moun-
tain more curls than those of their counterparts. tain were modest. First of all we wanted to check
A substantial difference concerns the execution of whether the information provided by Goell and
the Kommagenes: the west one is younger, more Brokamp was correct and to look if there was per-
elegant than the east one. In fact, that is true for haps more to be found. Second, we wanted a
the whole set of heads on the West Terrace. more thorough and detailed insight into the area.
It seems logical to surmise that the heads were Third, we looked for pottery and other traces of
made separately from the bodies, probably in a human occupation. The striking absence of pot-
workshop. The differences in style and execution tery on the terraces itself plays an important role
of the heads could be explained by assuming that in the discussion whether the sanctuary was ever
artisans with different skills executed the job in a in use.90 We hypothesized that the remains looked
large studio. It is unnecessary to postulate differ- for on the mountain might be, at least partly, in
ent data or different masters. the area below. This, firstly, in view of the clima-
The fact that one figure remained unfinished tological circumstances (every spring huge
may not puzzle us too much. Apparently the mon- amounts of melting water run down the moun-
ument had to be finished at a certain moment - the tain) and secondly from observations during the
king being old as he stated in his text - and one did preparation of the season (figs. 32-33).
care no longer for this shortcoming that would
remain practically invisible at the great distance. It 4.3.2 Results
is easy nowadays to study the details of the heads,
standing as they are on the two terraces, but it is As to our first question, the information provided
really doubtful whether one could distinguish all by Goell and Brokamp results to be correct and
their determining elements clearly from beneath. rather complete. The procession road stele I is still
in situ. There is no trace of the socket of stele II
4.3 The pilot survey (with Anne ten Brink) anymore; as Sanders already conjectured, the
place was probably destroyed while making the
As described more extensively in section 5, the foundations for the modern stairs leading up
area below tumulus and terraces was, by means from the cafetaria.91 From another stele, lying

104
below the North Terrace, it cannot be determined the water coming down can be clearly noticed by
anymore if it concerns an entrance stele or not the extreme weathering of the rock. The perennial
(see section 3.3). We did not come across other presence of water here is witnessed by vegetation
such remains. We would like to draw attention to at the borders of the caves still abundantly present
a flat area below the East Terrace where possible in August and September. Even in this period the
cuttings from the living rock are visible as well as guards working on the mountain fetch blocks of
large blocks of stone. There may have been a ice from the caves. We numbered the ice caves
quarry of some kind here. In general, the lime- directly connected with the gorges, but it is impor-
stone around the mountain gives us the impres- tant to realise that there were many more, stretch-
sion that the blocks for the statues could well ing out to the south and east. From the ice caves
have been hewn in the direct surroundings. it is a 20 minutes steep climb up to the East Terrace.
As to our second question, a more thorough The ice caves, providing abundant fresh and cold
inventarisation of the area proved worthwile. We water all year, thus seem to have been a source of
compiled a plan from the modern cafetaria water much nearer than the spring down the
(southwest) to the area below the East Terrace Malatya valley.
(northeast). The landscape consists for a large part Our search for remains of pottery and other
of an alternation of ridges and gorges. These human occupation, the third objective of the pilot
gorges are formed by the melting water running survey, proved to be succesful. In the survey area
down the mountain into one or more so-called ice we found prehistoric artefacts, pottery from later
caves. These ice caves form, as it were, a kind of periods, and parts of the monument on the
band at this side of the mountain; they continue mountain (stelae fragments, tuffit blocks, etc.). In
below the East Terrace and then stretch out fur- itself this is an important conclusion; however
ther east. The ice caves differ considerably in size. quantities are few, after difficult and intensive
In many, especially the larger ones, the power of searching. Most artefacts are found at the borders

Fig. 32. Nemrud Dağ, the survey area (2001, photo A. ten Brink).

105
 which different areas are statistically compared,
impossible.

4.4 Epigraphic research (Onno van Nijf & Marlies

Schipperheijn)

The archaeological research programme on

Nemrud Dag˘ includes an epigraphic component
under the responsibility of Onno M. van Nijf
(University of Groningen). During the campaign
of 2001 a small team from Groningen has made
an inventory of the epigraphic material still in
situ.93 This material can be divided into three
groups:
1. the so-called nomos inscription on the back
sides of the colossal statues on the East and
West terraces (figs. 7, 26-27)94;
2. the inscriptions on the back-sides of the dexio-
Fig. 33. Nemrud Dağ, survey find (2001, photo sis reliefs on the West terrace (fig. 34);
A. ten Brink). 3. a small number of fragments on an altar and
on scattered stones on the West terrace.
of the gorges. Remarkable is the presence of many
small iron slags. The survey database, to be com- The second group includes a number of palimpsest
pleted next years, will present the finds, supply- texts, i.e. inscriptions that have been chiseled over
ing a map of the area with the find numbers and already in antiquity. The original text is at places
further, for every artefact, hand-measured GPS still visible - if not readily legible.95 It has become
points, a description, dating and photographs. clear over the course of our investigations that
Regarding the prehistoric artefacts and the pot- these inscriptions have suffered severely from the
tery, the results are too meagre to draw real con- extreme weather conditions on the mountain, and
clusions at the moment, apart from the general from the large number of tourists who visit the
insight that Nemrud Dağ attracted human inter- site each year. Their condition is deteriorating
est from prehistoric times to the Byzantine rapidly (fig. 34).
period.92 During the campaign of 2001 we have taken
It is important to note that the artefacts that photographs and made paper squeezes of the
have fallen off the mountain cover a very large palimpsest inscriptions, and some of the others.
area. The few tuffit parts of stelae and sockets we We have focussed on the palimpsest texts, as it
found probably come from the East Terrace. They was expected that they contain important infor-
indicate that, again due to climatological circum- mation on the dating of the complex. Unfortu-
stances, the present findspots of artefacts fallen nately, we have not been able to make much
down from the terraces only have little to say progress with the decipherment, due to the bad
about their original position. quality of the text.
Finally, we were able to read two hitherto
4.3.3 Conclusion unnoticed inscriptions in Greek on the bottom of
the Herakles’ shoulder piece on the West Terrace.
A better knowledge about the area around the 1:
tumulus and terraces and the artefacts found Under the right arm two letters: C A.
there proves to be important for our understand- The sigma h. 3.5-3.8 cm, w. 2.0 cm, The alpha h.
ing of Antiochos’ monument and for the use of 3.5 cm, w. 2.5 cm.
Nemrud Dağ on a larger time scale. Further 2:
exploration is thus highly desirable. An extensive I AΛΛOΥ∆ΗΣΤΗΝΙΘΣ
field survey, however, seems impossible for prac- The total inscription is 54.0 cm long, the letters vary
tical reasons. The area is extremely rough and in height between 2.9 - 4.5 cm and in width between
there are considerable differences in landscape 2.0- 3.0 cm. Between the first I and the alpha is
types which, together with the extreme scarity of a space from 4.5 cm. The first I of the last word
the material, makes a more scientific survey in is doubtful, the theta may be read as an omikron.

106
 We have not been able to come up with a sat- is the surrounding landscape. Like the hundreds
isfying interpretation of the texts. It is important of tourists visiting the place every day at sunrise
to note that these inscriptions would not have and sunset, the ancient visitor must have been
been visible once the statue had been erected. The struck by this visual conquest of nature and will
quality of the letters is quite good, which suggests have gazed from the mountain in all directions.
that they may have been a sort of exercise of the To the North the Anti-Taurus continues and one
stonecutters. can see the Armenian Highlands; to the East the
The text may refer to a (local) name, but we Euphrates lingers, its valley being clearly visible,
have found no paralllels. If we have to read to the South where the plains of Syria begin. To
TENIOS, this might be an ethnic. We will com- the West, after a range of mountains, one can see
ment on this epigraphic material in the next valleys where, amongst other places, Arsameia
years. with the summer residence of the Kommagenean
The main aims of the epigraphic project for the kings is located as well as Karakus¸. As several
future are: times indicated in our description of the monu-
1. the decipherment and interpretation of the ment (section 3) this natural surrounding had, in
palimpsest texts; many ways, been made part of the concepts and
2. the collection of all the epigraphic material per- building of the hierothesion. This plan did not only
taining to Nemrud Dağ and the interpretation involve the monument proper: Nemrud Dag˘
of these documents within a wider cultural and was the centre of a much larger entity, a grid, as
historical framework that includes the other it were, with dynastic cult places on strategic
sanctuaries connected to the Kommagenian mountain tops in the surrounding, all with a clear
dynasty. view on Nemrud Dag˘. Also the area directly sur-
rounding the mountain was integrated in this
5. OUTLOOK FOR THE NEXT YEARS cultural landscape. As mentioned in the descrip-
tion (section 3) there were marked processional
One of the main characteristics of Antiochos’ roads leading to the sanctuary from different
hierothesion, and surely one of its most impressive, directions.

Fig. 34. Nemrud Dağ, West Terrace, back side of the reliefs with the palimpsest texts (2001, photo E.M.
Moormann).

107
 For the next years the Nemrud Dağ Project will There exists no comprehensive inventory of the
thus have three areas of interest. First and fore- artefacts from the monument, now scattered over
most the Nemrud Dağ hierothesion proper. Our the mountain itself, the villages around and the
main goal here is restoration and, at the same museums in Gaziantep, Adıyaman, Ankara, Berlin,
time, documentation (the SIS) and interpretation. etc. We will begin with defining the outlines of an
It is striking how few interpretative studies on the inventory incorporating all artefacts from
monument exist; it is only with the sound essay Nemrud Dağ.
of Wolfram Hoepfner that we begin to under- As far as epigraphy is concerned, there will be
stand the meaning of the place in a wider, made detailed 3D digital images of all the
Hellenistic context.96 inscribed surfaces on Nemrud Dağ, with special
The documentation work (the SIS) and inter- attention for the palimpsest texts. For this pur-
pretative efforts will continue as described for the pose a sophisticated 3D scanner has been put at
2001 campaign. Also some new projects will be our disposal by the Groningen Institute of
started. Most important are the beginning of a Archaeology.
structural engeneering project and a stone con- This large restoration, conservation and docu-
servation project. Both will be executed in coop- mentation project will be contextualised by the
eration with the World Monuments Fund (WMF). study of the area around the mountain, the sec-
For the 2002 campaign the structural enge- ond point of interest, as well as, thirdly, the
neering project aims at: Kommagenean landscape and archaeological
1. a stabilization of dislocated stones at both ter- remains in a wider sense.
races (specially at the East Terrace) and a
bedrock stabilization of the row of statues; NOTES
2. a reconstruction of the base of the lion/eagle
pair, in the first instance on the north side of 1 Composition of the team: professor Herman A.G.
the East terrace, which will be done by consol- Brijder, project manager, professor Eric M. Moormann
and dr Miguel John Versluys, acting project managers,
idation of the existing part with necessary drs Tesse D. Stek and Ellen Thiermann, SIS, all
repair and/or strengthening and re-building of University of Amsterdam; professor Onno M. van Nijf
damaged parts using dry stone masonry; and Marlies Schipperheijn, epigraphists, University of
3. anastilosis of the eagle statue on the prepared Groningen; ir. Hans Garlich and ing. Marinus Kremers,
geodesists, Technical University Delft; Anne ten Brink,
base. prehistorian, ir. Maurice L.A. Crijns, project coordina-
This first part of the engeneering activities tor, ir. Jaap Groot, constructing engineer, ing. Willian
should also be considered as a pilot project by A.M. van den Bogaard, Auto CAD, engineer and vir-
which methodology and techniques will be estab- tual reconstructions, drs. Petronella J.M. van den
lished and appropriate equipment will be checked Mortel, drawings and virtual reconstructions, all INF;
Dipl.-Ing. Klaus-Dieter Kiepsch, 3D laser-scan, Callidus
and selected for future conservation and stabi- Precision Systems, Halle (Saale); Nurhan Turan and
lization work. Also some emergency work will be Levent Vardar, representatives Ministry of Culture of
executed, like the stabilization of the bedrock Turkey. We would like to thank the Turkish Ministry of
where this is urgently needed (for instance under Culture for its kind cooperation, especially Dr. Alpay
Pasinli for his personal commitment to the project. Also
the Zeus and Antiochos statues on the East the great help of Halil Işik, governor of the province of
Terrace) Adıyaman, and of Bülent Akarcali is acknowledged
The stone conservation project will deal with with gratitude.
the condition survey of the monuments on the 2 Sanders 1996.
site, in situ testing of conservation materials and
3 See already the (critical) review by B. Jacobs, AM 30
(1998) 339-346.
procedures as well as sampling of stone materi- 4 See Sanders 1996, 26-32 for a more extensive overview
als for subsequent analysis and laboratory testing. of previous research.
This way we hope to be able to present a risk 5 D.J. van Lennep (1774-1853), Disputatio de regibus
assessment for the archaeological monument of Commagenes et Ciliciae Seleucidarum posteris, lecta D. XVII
mensis augusti et VIII mensis decembris MDCCCXXVIII,
Nemrud Dağ that will guide our further restora- no place, no date; Th. Mommsen, AM 1 (1876) 27-36.
tion and conservation activities. Both publications already mentioned by Wagner 2000,
The 2001 campaign established the fact that 11. Van Lennep presents an overview of the history of
there is indeed pottery to be found on the moun- the area and its kings from the beginning of the 3rd cen-
tain. We will therefore continue the survey and, tury BC onwards. In some cases coins are used as
pieces of evidence but the research is soundly histori-
moreover, try to come to a dating of our scarce cal and based on the study of ancient sources like
finds in cooperation with the Leiden Institute for Josephus’ Antiquitates and Dio’s Roman History. Van
Pottery Technology. Lennep even wrote another essay on this region:

108
 Disputatio de rege Bostreno aliisque rebus memoratis in 37 In Sanders 1996, 254-355 (both terraces). See on the
Epistula Ciceronis ad Q. fratrem lib. II, 12. Lecta in Instituti ancestor stelae most recently Messerschmidt W. 2000,
Regii Belgici classe tertia d. XV mensis Augusti MDCC- Die Ahnengalerie des Antiochos I. von Kommagene, in:
CXXXI (n.p., 14 p.) on the city of Zeugma. Gottkönige am Euphrat, 37-43 and Jacobs B. 2000, Die
6 For information on Dörner and an overview of his pub- Reliefs der Vorfahren des Antiochos I. von Kommagene
lications, see his Festschrift edited by S. Şahin, E. auf dem Nemrud Dağı - Versuch einer Neubenennung
Schwertheim, J. Wagner 1978, Studien zur Religion und der Frauendarstellungen in den mütterlichen Ahnen-
Kultur Kleinasiens, Leiden. The excavation of Arsameia, reihen, IstMit 50, 297-306.
all in all the most thoroughly investigated Kommagenean 38 Sanders 1996, 128.
site, was published in two volumes: Dörner/Goell 1963 39 Sanders 1996, 94.
and Hoepfner 1983. 40 For a short characterisation of both stone types see
7 See the list of expeditions in Sanders 1996, 82-85. Düppenbecker/Fitzner 1991. Sanders 1996 uses the
8 Waldmann H. 1973, Die kommagenischen Kultreformen term ‘sandstone’ throughout his work to describe the
unter König Mithridates I. Kallinikos und seinem Sohne ‘tuffit’. Sanders 1996, 156, note 21 vaguely hints at the
Antiochos I. , Leiden (EPRO 34); idem 1991, Der komma- place of the tuffit quarry. M. Crijns drew our attention
genische Mazdaismus, Berlin (IstMitt Beih. 37) (note how- to the spot.
ever that his conclusions are disputed and not gener- 41 Some cuttings in this small mountain suggest the pres-
ally accepted); Şahin 1991. ence of rooms, probably for officials or guards.
9 Wagner 2000, with previous literature. 42 Sanders 1996, 192, fig. 129.
10 Sanders 1996, XVII: ‘… it is clear that Goell and her 43 See Sanders 1996, fig. 100. Diameter shoulder socket 54
assistants moved or at least overturned nearly every cm, diameter tenon 48 cm.
significant stone on the site. Thus, the current position 44 This peculiarity is also observed in Sanders 1996, 420,
and associations of such stones are not reliable guides fig. 165. A similar incision can be seen on a block of the
to original site conditions.’ left eagle on the East Terrace. Other differences between
11 The best reconstruction is Utecht/Lütjen/Stein 1993, the colossi of both terraces are listed in Sanders 1996,
esp. fig. 5. ibidem 449-454. He concludes (p. 454) that the colossi
12 Such a bank made with large blocks is visible behind are identical despite these small differences.
the colossi on the East Terrace. 45 To give some examples, on the West Terrace, the heads
13 See Utecht/Lütjen/Stein 1993. of Zeus, Kommagene and, Antiochos have four holes,
14 Hoepfner 2000, 65. Apollo has two holes, the eagle has three holes (the
15 To indicate them, we will further only use their first (i.e. southern one has the beginning of a fourth at one side).
Greek) designation. 46 As a parallel one can point at the reconstruction project
16 The right foot of Zeus’ throne is resting on some tuffit recently undertaken in the sanctuary in Xanthos (Lycia,
slabs. Turkey) where the temple of Leto is being reconstructed
17 Sanders 1996, 392 sees them as attributes used during with 80% of the blocks still present, cf. Archéologia
prayers and offerings. In that case our figures are 1/2002, 15 ff.
‘spendende Götter’. 47 This work was carried out by members of the Technical
18 Sanders 1996, 158. University Delft, cf. for a detailed account Garlich
19 Sanders 1996, 248, 448-449 (coronation), 252-253 (frag- J.H./M.J.M. Kremers, GPS-RTK mapping of Nemrud
ments of Lion Horoscope, including the stars’ names). monument, Geodesia 2002-5, 188-193.
On the plateau see Sanders 1996, 106-109. 48 This work has not been done yet; finishing SIS level 2
20 Sanders 1996, 104-105. for the East Terrace is one of the goals of the 2002 cam-
21 Sanders 1996, 159 n. 37. paign (see further section 5).
22 For a definition of tuffit cf. section 3.4. 49 See for this aspect in general Beck R. 1962, Epiphanes of
23 Sanders 1996, 230-231, fig. 266-267. the great gods: astrology in the royal cult of Commagene,
24 Sanders 1996, 123, 230. Toronto.
25 In the following we describe them as if they were com- 50 The following experts helped to tackle the problem of
pletely preserved. the Lion Horoscope: J. Meeus, J. Rademaker, C. de Jager
26 Sanders 1996, 397-398. (Space Research Organization, Utrecht), and G.
27 This practice is clearly visible on the head seen in pro- Schilling (Artis Planetarium, Amsterdam).
file of the king on the dexiosis stele with Apollo. 51 C. Robert, Eratosthenis catasterismorum libri XII (1878)
28 This detail is hardly recognisable, but was observed esp. 96.
and analysed by Goell (Sanders 1996, 200, 388, 470). 52 Humann/Puchstein 1890, 331-336.
29 Re-use was already observed by Goell: Sanders 1996, 53 Humann/Puchstein 1890, 330.
121, 232, 254, see further our section 4.4. 54 Still starting from the premise that the date had to be
30 The stelae have been erected by Dörner in 1984 (Sanders found within the first half of the 1st century BC. Cf.
1996, 163 note 66). The base was slightly restored: the Neugebauer O./H.B. van Hoesen, 1959, Greek Horoscopes
original situation can be seen on Sanders 1996, fig. 304. (= Memoirs of the American Philosophical Society, 48),
31 For a detailed description see Sanders 1996, 232-254. Philadelphia, esp. 14-16 cat. no. 61.
32 Sanders 1996, 396, 401, 410-413, only pointing at this 55 Sanders 1996, 172.
feature specifically on page 445. 56 W. Haase, AW Sondernummer 6, Kommagene (1975) 21.
33 Sanders 1996, 121, 248 fig. 272-277. Cf. p. 234. 57 Tuckerman B., 1962, Planetary, Lunar and Solar Positions
34 Goell feels a sense of humour in this detail and com- 601 B.C. to A.D. 1, Philadelphia.
pares the ears to those of rabbits (Sanders 1996, 242-243). 58 O. Neugebauer had calculated this date as a possible
35 Sanders 1996, 244. option.
36 Sanders 1996, 306: Goell replaced the altars; 355 59 The sequence of these heavenly bodies as depicted on
(inscription). the Lion horoscope is different (from right to left:

109
 Moon, Jupiter, Mercury and Mars) due to the differing 85 Antiochos helped Pompey with 200 archers in his bat-
velocities of their circulation. tle against Caesar (Caesar, Bell. Civ. III 4; Appian, Bell.
60 Humann/Puchstein 1890, 346. Civ. II 49), cf. Sanders 1996, 23; Wagner 2000, 25 (time
61 A handy chronological table can be consulted in table).
Wagner 2000, 25. The last date differs: one also finds 38 86 Hoepfner 2000, 68 considers the choice as free, done at
and 32. random. He follows the idea of a pantheion as proposed
62 East IA, 20-24; text in Sanders 1996, 208. in older publications.
63 Young 1964 (heads); Şahin 1991, 116 (utensils); Hoepfner 87 Young 1964. We seriously consider the possibility that
2000, 66; B. Jacobs, IstMit 50 (2000) 303. the Apollo (fig. 11) was left unfinished as well because
64 Goell (in Sanders 1996, 232) thought that Antiochos of the extremely roundish outlines and the fish-like
probably had planned a nomos text, later replaced by closed lips. Whether true or not, it doesn’t detract from
the text we see nowadays. No chronological conse- the general conclusions.
quences were drawn from this observation. 88 This peculiarity is also observed in Sanders 1996, 453,
65 No stele portraying Antiochos himself has been found. but without drawing any conclusions out of it.
According to Jacobs 2000, 48-49 Antiochos did not erect 89 The pilot-survey was planned, set-up and executed by
an image of himself in this series because he was placed Anne ten Brink and Miguel John Versluys.
among the gods. Notwithstanding Jacobs’ dating of the 90 See, for instance, Şahin 1991.
whole monument to the reign of Antiochos (p. 34-35), 91 Sanders 1996, 153 note 9. On the entries and the stelae
this may be true. Sanders 1996, 94-96.
66 The work can have been done in a rather short lapse of 92 On the terraces Goell found in all years of excavation
time. Utecht/Lütjen/Stein 1993 point at the relatively some six Roman and Byzantine coins, see Sanders 1996,
small amount of energy and work needed to construct 472-474.
the tumulus with its layer of gravel. 93 Remains and fragments of inscriptions that were dis-
67 See most recently Wagner 2000, 18, 21-22. covered by earlier campaigns are now dispersed over
68 Sanders 1996, 449. a number of Turkish and other musea.
69 Wagner 2000, 23-24 points at the simple forms of the 94 Publications: Humann/Puchstein 1890, 262-278;
Karakuş tumulus. Cumont II 1896, 89-91; Michel 1900, 735; OGIS 383, 591-
70 Waldmann 1973 (see n. 8), 54; Şahin 1991, 115. 603; Laum 1914, 148-153; Waldis 1920, 3-11; IGLS I,
71 Sanders 1996, 141-143 and 163 n. 65; Jacobs 2000, 35; 1929, 12-26; Dörner 1953, 143-146; Waldmann 1973, 61-
Hoepfner 2000, 66. 71; Sanders 1996, 207-213.
72 Hoepfner 2000, 67. 95 Puchstein already noticed this palimpsest as is clear
73 Hoepfner 2000, 65 considers these steps as ‘Sitzstufen from a footnote in description of the Zeus-stele: ‘Unter
für Zuschauer in der Art einer Theateranlage’. Even if dieser grossen Weihinschrift habe ich bei gewisser
the idea is plausible, one may wonder that the atten- Beleuchtung auf der glatt gewaschenen Fläche des
dants could not see the statues which played a funda- Steins bisweilen lange Reihen kleiner Buchstaben
mental role in the setting (and perhaps in the cere- wahrzunehmen geglaubt, ohne dass es mir möglich
monies to be celebrated). war, irgend einen bestimmten Buchstaben zu erkennen.
74 Sanders 1996, 106: ‘a row of five sandstone reliefs Auf einem Abklatsch dieser Fläche is dergleichen
flanked at the north and south ends by a group com- jedoch nichts zu bemerken.’ (Humann/Puchstein 1890,
posed of free-standing lions and eagles’. 325).
75 See Sanders 1996, figs. 318-322 (Lion horoscope frag- 96 Hoepfner 2000.
ments) and 234-235 (other fragments).
76 Humann/Puchstein 1890, plate XXII. BIBLIOGRAPHY/ABBREVIATIONS
77 According to one of our local informants, who prac-
ticed the catchment of kekliks, these birds are catched Dörner, F.K./T.B. Goell 1963, Die Ausgrabungen im
by luring birds to another put into a cave and there is Hierothesion des Mithridates Kallinikos von 1953-1956
no need of a large built construction. (IstForsch 23), Berlin.
78 Hoepfner 2000, 65: ‘Zentrum der Kulthandlung ist in Düppenbecker, V./B. Fitzner 1991, Gesteineigenschaften
jedem Heiligtum ein Altar, auf dem der große Opfer und Verwitterungszustand der Monumente auf dem
vollzogen wurde.…’ But the idea of a ‘fire altar’ is Nemrud Dağ, in Nemrud Dağ - Neue Methoden der
‘Folge einer willkürlichen Bezeichnung’. The partici- Archäologie, Bönen, no page numbers.
pants were to sit on the plateau under the statues. Gottkönige am Euphrat = Wagner, J. (Hrsg.) 2000, Gottkönige
79 Utecht/Lütjen/Stein 1993, fig. 14, no. R2. am Euphrat. Neue Ausgrabungen und Forschungen in
80 Apparently Hoepfner 2000, 70-71 does not object to the Kommagene, Mainz am Rhein.
concept of an altar, comparing it to a pantheon at Kamiros. Hoepfner, W. 1983, Das Hierothesion des Königs Mithridates
81 This element seems not to have been changed during I. Kallinikos von Kommagene nach den Ausgrabungen von
the restoration work by Dörner in 1984 (see Sanders 1963 bis 1967 (IstForsch 33), Berlin.
1996, fig. 271), as becomes clear from Sanders 1996, fig. Hoepfner, W. 2000, Arsameia am Nymphaios und der
304. Sanders 1996, 163 note 66 tells that Dörner even Allgötterkult Antiochos I. Schriften, Bilder und Säulen
had tuffit blocks made for the basis. For Dörner’s als Zeugnisse späthellenistischer Kultur, in Gottkönige
restoration see Sanders 1996, fig. 271. am Euphrat, 56-73.
82 Humann/Puchstein 1890, 282. Humann, K./O. Puchstein 1890, Reisen in Kleinasien und
83 The drawings, however, show the old sequence (e.g. Nordsyrien, Berlin.
fig. 87), which makes the reading of Goell’s work rather Jacobs, B. 2000, Das Heiligtum auf dem Nemrud Daği. Zur
confusing. Baupolitik des Antiochos I. von Kommagene und
84 A fragment of the tip of Antiochos’ tiara of the East seines Sohnes Mithridates II., in Gottkönige am Euphrat,
Terrace lies near the Kommagene head. 26-35.

110
Şahin, S. 1991, Forschungen in Kommagene, Epigraphica Young, J. H. 1964, Commagenian Tiaras: Royal and Divine,
Anatolica 18, 99-132. AJA 68, 29-34.
Sanders, D.H. (ed.) 1996, Nemrud Daği. The Hierothesion of Wagner, J. 2000, Die Könige von Kommagene und ihr
Antiochos I of Commagene. Results of the American Herrscherkult, in Gottkönige am Euphrat, 11-25.
Excavations directed by Theresa B. Goell, I (text), II (illus-
trations), Indiana. AMSTERDAMS ARCHEOLOGISCH CENTRUM
Utecht, T./H. Lütjen/B. Stein 1993, A Short Note on the
NIEUWE PRINSENGRACHT 130
Geophysical Survey of 1989 at the Archaeological Site
of Nemrud Dagh, Turkey, in Vogel A./G.N. Tsokas NL-1018 VZ AMSTERDAM
(eds.), Geophysical Exploration of Archaeological Sites,
Wiesbaden, 24 pp., no page numbers.

111