# STUDIA ASIANA

$$-9-$$

# STUDIA ASIANA Collana fondata da Alfonso Archi, Onofrio Carruba e Franca Pecchioli Daddi

#### *Comitato Scientifico*

Alfonso Archi, Fondazione OrMe – Oriente Mediterraneo Amalia Catagnoti, Università degli Studi di Firenze Anacleto D'Agostino, Università di Pisa Rita Francia, Sapienza – Università di Roma Gianni Marchesi, Alma Mater Studiorum – Università di Bologna Stefania Mazzoni, Università degli Studi di Firenze Valentina Orsi, Università degli Studi di Firenze Marina Pucci, Università degli Studi di Firenze Elena Rova, Università Ca' Foscari – Venezia Giulia Torri, Università degli Studi di Firenze

# Sacred Landscapes of Hittites and Luwians

Proceedings of the International Conference in Honour of Franca Pecchioli Daddi Florence, February 6th-8th 2014

Edited by Anacleto D'Agostino, Valentina Orsi, Giulia Torri

firenze university press

2015

Sacred Landscapes of Hittites and Luwians : proceedings of the International Conference in Honour of Franca Pecchioli Daddi : Florence, February 6th-8th 2014 / edited by Anacleto D'Agostino, Valentina Orsi, Giulia Torri. – Firenze : Firenze University Press, 2015. (Studia Asiana ; 9)

http://digital.casalini.it/9788866559047

ISBN 978-88-6655-903-0 (print) ISBN 978-88-6655-904-7 (online)

Graphic design: Alberto Pizarro Fernández, Pagina Maestra

Front cover photo: Drawing of the rock reliefs at Yazılıkaya (Charles Texier, *Description de l'Asie Mineure faite par ordre du Governement français de 1833 à 1837*. Typ. de Firmin Didot frères, Paris 1839, planche 72).

The volume was published with the contribution of *Ente Cassa di Risparmio di Firenze.*

#### *Peer Review Process*

All publications are submitted to an external refereeing process under the responsibility of the FUP Editorial Board and the Scientific Committees of the individual series. The works published in the FUP catalogue are evaluated and approved by the Editorial Board of the publishing house. For a more detailed description of the refereeing process we refer to the official documents published on the website and in the online catalogue of the FUP (www.fupress.com).

#### *Firenze University Press Editorial Board*

G. Nigro (Co-ordinator), M.T. Bartoli, M. Boddi, R. Casalbuoni, C. Ciappei, R. Del Punta, A. Dolfi, V. Fargion, S. Ferrone, M. Garzaniti, P. Guarnieri, A. Mariani, M. Marini, A. Novelli, M.C. Torricelli, M. Verga, A. Zorzi.

La presente opera è rilasciata nei termini della licenza Creative Commons Attribution 4.0 Unported (CC BY 4.0: https://creativecommons.org/licenses/by/4.0/legalcode).

**CC** 2015 Firenze University Press Università degli Studi di Firenze Firenze University Press Borgo degli Albizi, 28 50122 www.fupress.com *Printed in Italy*

# CONTENTS


# FOREWORD

# *Anacleto D'Agostino, Valentina Orsi, Giulia Torri*

On February 6th-8th 2014 the Department of History, Archaeology, Geography, Fine and Performative Arts (SAGAS) of the University of Florence hosted an international conference, *Sacred Landscapes of Hittites and Luwians*, which brought together scholars from different areas of Anatolian Studies who shared an interest in various aspects of the symbolic landscapes built by the Hittite and post-Hittite societies respectively in the Second and First millennia BC.

We take great pleasure in presenting the proceedings of this conference in the hope that the contributions published here will find the appreciation of the scholarly community and stimulate debate on an important issue of the religious landscape of the Anatolia communities between the Late Bronze Age and Iron Age. We express our gratitude to the colleagues who participated in the conference and those who have contributed to this book.

Our intent was to offer an opportunity to discuss the significance of the intertwining of landscapes, architectures and topographies, which has recently become a very debated topic. The main purpose was to investigate how Hittite and other Anatolian populations represented and built their sacred landscapes in the course of the centuries.

Archaeologists, hittitologists and historians stressed how the populations of the plateau perceived many tangible and intangible elements of the Anatolian environment, like mountains, rivers and rocks, but also atmospheric agents, and natural phenomena as parts of a symbolic construction of the *sacrum*. Human communities transformed Anatolian landscape over the centuries, interpreting it as a natural and essential part of their religious and ideological world. By altering the features of this landscape, performing religious and social actions and reshaping the countryside with their techniques, they created a unity between human beings and Nature, expressing in this way their identity. From the articles published here it emerges that the natural open-air places, rock peaks, springs, as well as cities, buildings and gates or certain portions of the settlements had their own specific sacredness, where ritual practices were enacted. Numerous testimonies of such a state of affairs are contained in Hittite cuneiform scripts but also in the millenary monuments on the Anatolian landscape which are often accompanied by Hieroglyphic inscriptions.

Twenty-seven papers were presented and a large public of scholars and students attended the lectures in the Aula Magna of the Palazzo del Rettorato dell'Università degli Studi di Firenze, in Piazza San Marco, and in the halls of Piazza Brunelleschi (Sala Comparetti) and via Laura (Aula 6). Thirteen colleagues sent their contributions for the present volume.

In addition, a session of works was dedicated to the presentation of recent research. This offered the opportunity to discuss the results of archaeological projects and studies in progress carried out by Italian scholars in the sites of Uşaklı Höyük, Karkemiš and Zincirli.

The conference was conceived by Stefania Mazzoni and Franca Pecchioli Daddi and organized by Carlo Corti, Anacleto D'Agostino, Valentina Orsi and Giulia Torri as a part of a larger project funded by the Italian government on the strategic project (PRIN 2009) *Modelli di costruzione fisica e ideologica del territorio e identità culturali: città sacre, santuari, complessi funerari in Siria, Anatolia e Transcaucasia nelle Età del Bronzo e del Ferro*. Silvana Rubanu worked at the registration desk and as assistant in the halls where the papers were presented.

This publication was funded by the Fondazione Ente Cassa di Risparmio di Firenze as a part of the financed project *Il paesaggio sacro della Siria fra II e I millennio a.C.*

The works of the conference were opened by Alberto Tesi, then Dean of the University of Florence, and Anna Benvenuti, Director of the Department SAGAS, to whom goes our gratitude for their support.

The conference was dedicated to Franca Pecchioli Daddi who was to retire from the academic service at the University of Florence the same year. Franca passed away too soon after this event but the words of Alfonso Archi addressed to her on the occasion of the conference, which now open this volume, are the best way to remember her.

This book is dedicated to her memory by all the contributors of the volume.

# PER FRANCA

# *Alfonso Archi*

#### Dear Franca,

with this academic year you will end your work directly with this university and set off on that marvellous period of your life when you can dedicate yourself entirely to your own research. In fact, those of us working in the humanities have the great good fortune of this privilege: to continue our studies for as long as we wish. For the wise, this is a magical moment, I can assure you.

You will find yourself at the end of the long journey of your university life. Our shared Maestri in this seat of learning: the Università di Firenze, great Masters who had to face up to the challanges of one of the most dramatic periods in our country's history: the years of the reconstruction after the war, taught you that it is also essential to strive to improve the institutions in which we find ourselves working. Thus, it was a duty for you to take part in the radical task of transformation that has changed our universities and made them capable of satisfying new needs and requirements in the field of higher education. It was a lengthy period, full of endless discussions, proposals and deliberations which were immediately overturned by new ministerial decisions, a continuous making and undoing which, at a certain point, dulled the enthusiasm and hopes of many, hopes which were, sometimes, objectively unrealistic. A difficult period which lasted too long and which now would seem to have come to an end, and one to which you sacrificed much of your time.

They were times in which a professor close to retirement was not considered as a person but merely an entry in the budget, as was noted some years ago by an illuminate intellectual as Claudio Magris, former professor of German Literature in Trieste. In such circumstances it was necessary to hold firm and steer to course of what, in simple terms, is defined a cultural policy: knowing how to reconcile new didactic needs with the cultural lines that characterise a given Faculty. And this you did, choosing to become Dean of this Faculty for a number of years.

The chair of Storia dell'Asia Anteriore Antica of Florence University was instituted in 1954 and the post assigned to our Maestro, Giovanni Pugliese Carratelli, whose *curriculum vitae* was anything that usual. Author of the *editio princeps* of tablets in Linear A from Haghia Triada; of an edition of Greek inscriptions from the Dodecanese, he was also contributing intensively to the studies in Linear B, immediately after its deciphering by Michael Ventris, and had edited the *Tabulae Herculanenses*. He was a great humanist and, as a consequence, had pupils also from outside the field of Classical History. When he left Florence, the torch of Anatolian studies was taken up by his pupil Fiorella Imparati (with a determination which, on occasion, only a woman can show). In 1984, thanks to Guido Clemente, at that time Dean of the Faculty, the chair of Archaeology of the Near East was created and the professorship given to another of Pugliese's students, Paolo Emilio Pecorella, who died tragically in his excavation.

Anatolian Studies, Hittitology, is certainly a 'small' discipline (one of those which in Germany are called *Orchideen Fächer*). It can contribute, however, to characterize in part a university institution, if conducted with continuity and produces results. This is a matter of fact, although sometimes it is difficult to convince some colleagues. When we visit a museum, however, we may see on the labels: such and such excavation, such and such university.

You, Franca, full aware of this, have also promoted field activity in Anatolia. Several years ago, we went, together with Anna Maria Polvani and Carlo Corti, in search of an excavation in Turkey, visiting Oymaağaç Höyük (the ancient Hittite sanctuary of Nerik). The time was not yet ripe in Florence, and our German colleagues had better chances in Ankara. It was you who later suggested going to Ușaklı Höyük, identified by Oliver Gurney as ancient Zippalanda, a sanctuary with even

#### X *Alfonso Archi*

more continuity than Nerik. I am convinced that Gurney was right: in any case, surveys and the first excavations have already revealed the presence there of Hittite monumental buildings and the existence of an archive.

Therefore, Franca, many thanks for your persistence in planning this exciting adventure, and for your enthusiasm in all your researching and teach. Those attending this meeting wish to dedicate their contributions to you.

Florence, 6th February 2014

# HITTITE MONUMENTS AND THEIR SANCTITY

*John David Hawkins*

#### *Abstract*

The greater part of the landscape of Hittite Anatolia was sacred, in that its mountains, rivers and springs were so regarded and invoked. In some places, Hittite monuments have been found which mark more clearly such sacred sites, e.g. Eflatun Pınar, Gavurkale, Sipylos-Akpınar. If, as often, these are accompanied by a Hieroglyphic inscription, this may define even better the character of the site, and perhaps reveal its ancient name. As I have argued, a series of monuments mark the establishment by Tutḫaliya IV on locations of his royal hunts of cult centres of the Stag-God and his consort Ala, and juxtaposing these with Tutḫaliya's festival for all the names of the Stag-God and Ala, we may locate the mountains Sarpa, Sarissa, Sunnara and perhaps Alatarma. The inscription KARADAĞ 1 marks a mountain-top shrine of the divine 'Great Mountain'. The line of monuments Fraktin, Taşçı, İmamkulu and Hanyeri lay on or near a route through a mountain pass, Gezbel, and each no doubt had its own peculiar sanctity, most obviously in the case of Hanyeri, which depicts and names a sacred mountain and its patron deity Šarruma 'mountain king'. At Yalburt, a historical inscription recording a Lukka campaign of Tutḫaliya IV was incorporated into a pool structure, presumably at a sacred spring. Some sites marked by monuments may be both sacred and also serve as frontier markers, thus KARABEL on the pass between Mira and the Seḫa River land, and HATIP at the beautiful spring between Tarḫuntassa and Hatti. Other inscriptions however appear to be more like graffiti recording the names of passersby: SIPYLOS, MALKAYA and LATMOS.

Other such monuments belong to the post-Empire period. The remarkable hydronomy of the İvriz area could not but attract sanctity, signalled by the long-known great relief and the recently added stele and colossal statue fragment. Other inscriptions seem to record particular events without obvious sanctity: TOPADA (a battle), GÜRÜN (establishment of a city's territory), KÖTÜKALE (a road building), SUVASA (graffiti).

One further feature of the sacred landscape deserves a new look: the DKASKAL.KUR, brilliantly elucidated nearly 50 years ago by Edmund Gordon as 'sink hole, underground water course', Turkish *düden*.

#### *1. Introduction*

The greater part of the landscape of Hittite Anatolia was in some sense sacred, in that the mountains, rivers and springs were so regarded, and were invoked after the gods themselves as witnesses to treaties. Similarly, Muwatalli II in his prayer to the *piḫassassi* Storm-God not only includes the mountains and the rivers along with the gods of most localities invoked but also on occasion names them, mountains more than rivers. Divine mountain-men regularly appear in artistic representation of reliefs and seals, also as figurines. While mountain names in cuneiform texts are not normally given the divine determinative DINGIR, and only rarely in Hieroglyphic (Mount Sarpa on EMIRGAZI altars, Mount SWORD on HANYERI, GREAT.MOUNTAIN on KARADAĞ), this does not diminish their obvious sanctity. In this context a topographical feature which is marked as divine, the DKASKAL. KUR, will also be considered in this paper.

The Anatolian landscape is also dotted which man-made monuments attaching to various features, which bring with them indications of sanctity. When uninscribed, the character and the purpose of

#### 2 *JOHN DAVID HAWKINS*

such monuments may or may not be obvious. In such cases there has often been much discussion as to whether these can be recognized as commemorative or funerary, and in particular whether they may be identified with terms appearing in cuneiform texts, the «peak sanctuary» (NA4*ḫekur*) or the «Stone House» (É.NA4 ). For a recent evaluation of these terms, see van den Hout 2002, «Mortuary chapel» and «Mausoleum, tomb»; also HW2 b III/2, lief. 8, 2010, s.v. (NA4)*ḫekur*.

When however these monuments are accompanied by an inscription, this may give a clue or even direct information about the monument itself. My purpose in this paper is to review what these inscriptions may tell us. First, however, let us consider several uninscribed monuments which have been identified with terms occurring in the texts.

#### *2. Gavurkalesi*<sup>1</sup>

This rocky hill-top with rock-cut sculptures, ruins of a cyclopean precinct wall and a stone-built chamber presents features which we might expect both from a 'stone-house' tomb and a *ḫekur* mortuary precinct. The main problem in interpreting the site is the question whether such an elaborate monument could have been constructed for any individual other that royalty, set against an assumption that Hittite royalty might have required burial and mortuary cult to be located in Ḫattusa itself rather than in a distant site in the middle of the countryside. So if not royalty, then who?

#### *3. Eflatun Pınar*<sup>2</sup>

Up from the north-east side of Lake Beyşehir west of Konya lies the copious perennial spring Eflatun Pınar celebrated by the famous Hittite monument, a stone-built structure on the edge of an artificial pool. Discovered as long ago as 1837 (Hamilton 1842), the monument and its iconography and purpose have been much discussed. In the absence of any identifying inscription, its date and builder are uncertain, though the execution and the style point to the later Hittite Empire, 13th century BC, and an attribution to Tutḫaliya IV is not improbable. From 1999 archaeological investigations and the dredging of the pool revealed the long-hidden lower course of the structure's façade, a row of five mountain-men wearing pointed helmets and skirts pierced with holes to jet water.

It can hardly be doubted that the monument is connected with the sanctity of the spring.

#### *4. Sipylos-Akpınar*<sup>3</sup>

The 'Mother-goddess' figure on the north side of Manisa Dağ (classical Mount Sipylos) is clearly visible from the plain below and has been known since classical antiquity. It has been visited and reported by travellers at least since the 18th century. The much eroded carving in its arched niche has been variously described: some have seen it as a seated female figure, thus 'Mother-goddess', others have seen what appear to be traces of a beard, thus a male figure, also perhaps standing. A link between the figure and the abundant springs at the foot of the hillside has been suggested, though no ancient construction around these has been reported. The figure itself, whatever it represents, can hardly be other than divine, and the spring themselves are also such as would attract a notion of sanctity, so the idea that the two together represent a sacred precinct is not unreasonable.

Attached to the figure are two small Hieroglyphic inscriptions representing names. The first is placed high up on the right side of the niche, a rectangular incised panel, its component signs in raised relief, reading (sinistroverse) «Kuwalanamuwa, prince». The same name and title are attached to the figures with bow and spear on the İMAMKULU and HANYERİ monuments, for which see below. The other inscription, incised, further to the rght and lower, reads (sinistroverse) «Zuwani, eunuch, …» (further title, uncertain reading). This incised inscription has the appearance of a graffito left by a casual visitor.

<sup>1</sup> Kohlmeyer 1983, no. 6, pp. 43-48, Taf. 18-20. Rossner 1988, no. 4, pp. 57-63. Ehringhaus 2005, 11-14.

<sup>2</sup> Kohlmeyer 1983, n. 5, pp. 34-43, Taf. 12-15; Rossner 1988, n. 6, pp. 67-74; Emre 2002, p. 222, 228, 230; Ehringhaus 2005, pp. 50-57.

<sup>3</sup> Kohlmeyer 1983, n. 4, 28-34; Rossner 1989, n. 1, 39-45; Ehringhaus 2005, 84-87.

#### HITTITE MONUMENTS AND THEIR SANCTITY 3

The first inscription on the other hand, being in relief, was more carefully and laboriously executed. It could, but need not, be connected with the creation of the divine figure. Was it intended as the signature of the author of the monument? And could he have been the same individual as the prince(s) named on İMAMKULU and HANYERİ? These are possibilities but are by no means certain. The only Kuwalanamuwa attested in Hittite texts was a commander [of something] named in his annals by Mursili II in the period before his accession (Götze 1933: 26). The bigger question is whether the monument is the work of the Hittites, thus possibly executed on a Hittite western campaign, or whether perhaps it is genuinely Arzawean executed by a king of Arzawa or of one of 13th century Arzawa lands. The presence of the nearby KARABEL relief, work of a king of Mira at least suggests this possibility.

#### *5. Sirkeli*<sup>4</sup>

The earliest securely datable figure of a Hittite king, also the earliest Hieroglyphic monumental epigraph, are provided by the representation of Muwatalli II, identified by Hieroglyphic epigraph, on a cliff on the south bank of the river Ceyhan, where it breaks through the mountainous area (Misis Dağları) in the middle of the Cilician plain. Known since 1934-1937, the relief has attracted renewed attention with the publication in 1988 of the Bronze Tablet treaty of Tutḫaliya IV with his cousin Kuruntiya, king of Tarḫuntassa, one clause of which (§10) relates to Kuruntiya's right to access to the «Eternal peak sanctuary» (NA4*ḫekur* SAG.UŠ). Although not so stated, it has been supposed that this installation recognized as mortuary chapel was that of Kuruntiya's father Muwatalli II, and further that Sirkeli relief might indicate its location at this site. Two rounds of excavation in the large *höyük* lying above the rock-face with the relief have been conducted (1992-1997 and 2006 onwards) without very conclusive results, so this interpretation of the building(s) excavated at Sirkeli remains no more that a possibility.

#### *6. Fraktin and Tașçı*<sup>5</sup>

The line of inscribed reliefs south of Kayseri and Erciyes Dağ, FRAKTİN, TAȘÇI, İMAMKULU, and HANYERİ appears to mark one of the routes from the Anatolian plateau to the Cilician plain, crossing the Zamantı Su and the Tahtalı Dağları by the Gezbel pass. The first two are explicitly associated with Ḫattusili III (himself and his queen, then his servants, all identified by Hieroglyphic epigraphs); the second two with the prince Kuwalanamuwa. The FRAKTİN relief, showing Ḫattusili and queen Puduḫepa (with extended title *daughter of Kizzuwatna, beloved of gods*), libating respectively to a tutelary deity and to Ḫebat, presumably indicates some sort of sanctity at the site. Attention has been drawn to 'cup marks' for libations cut into the rock above the relief figures, also to the abundant water, a small tributary of the Zamantı Su flowing through the fields at the foot of the low cliff. A large *höyük* in the vicinity has also been noted and subjected to preliminary investigations. But the nature of the sites sanctity is not explicitly indicated. TAŞÇI with its procession of servants of Ḫattusili, also a further solitary figure, is also situated directly on another tributary of the Zamantı Su. Strange, artificially enlarged hollows in the rock behind both the procession and the solitary figure suggest some kind of cultic function which is not more clearly indicated.

### *7. İmamkulu and Hanyeri*<sup>6</sup>

The İMAMKULU relief is placed on a relative undistinguished rock on a relatively undistinguished hillside overlooking the Zamantı Su to the west. The tripartite scene comprises: (1, left) a figure with bow and spear; (2, centre) the Storm-God driving his bull-drawn eagle chariot over the shouldres of three bowing mountain men, themselves supported by lion-headed (?) human figures; (3, right) a winged naked goddess above a composite beast. Left and centre figures are identified by Hieroglyphic epigraphs:

<sup>4</sup> Kohlmeyer 1983, no. 14, 95-101, Taf. 37-39; Rossner 1988, no. 31, 223-227; Ehringhaus 2005, 95-101, Hrouda 2011, s.v. Sirkeli.

<sup>5</sup> Kohlmeyer 1983, nos. 8-10, 67-80, Taf. 22-28; Rossner 1988, nos. 20-21, 159-172; Ehringhaus 2005, 69-70.

<sup>6</sup> Kohlmeyer 1983, nos. 11,12, 80-90, Taf. 29-33.; Rossner 1988, nos. 22-23, 173-185; Ehringhaus 2005, 70-80.

#### 4 *JOHN DAVID HAWKINS*

«Prince Kuwalanamuwa» (Hawkins 1983, s.v. Kuwatna-muwa); and «Storm-God of Halab» (Hawkins 2003). The right hand group has been plausibly identified on the basis of Hittite *Bildbeschreibungen* as representing the goddess Šauška with an *awiti*-monster, lion-headed and multiple-winged (Hazenbos 2002). It is not clear whether the centre and right-hand deities are contextually connected: another plausible suggestion has the bird between them as a dove carrying an erotic message (Keel 1992).

The figure of the prince Kuwalanamuwa appears again on the HANYERİ relief above İMAMKULU just the other side of the Gezbel pass. There the figure with bow and spear, again identified by a Hieroglyphic epigraph above his right arm, just in front of this face, faces left towards a right-facing bull, its fore-legs on the shoulder of a mountain-man, rear legs on a pedestal. This group too is identified by Hieroglyphic epigraphs: «King of the Mountain Šarruma» (the bull, Laroche 1963); «SWORD, divine mountain» (the mountain-man, Laroche 1969: 81). This inscription expresses much more clearly than others the locality's sanctity: the Mountain-King Šarruma represented as a bull, on the divine mountain named SWORD, presumably to be identified as the area's highest peak, the Bey Dağ, on the flank of which the relief is placed. The divine SWORD (Yazılıkaya no. 27) is identified as Cuneiform d U.GUR / Nergal, Hatt.-Hitt. Sulinkatte, Ugaritic Rašap, Hurrian Iršappa. The divine mountain's name is to be read in one of these languages, presumably Hurrian. The relief seems to show that this mountain is a/ the seat of the god Šarruma.

Behind the figure with bow and spear another prince's name is inscribed, to be read «Prince Tarḫuntami» (Hawkins 2005: 273; Dinçol and Dinçol 2008: 67). This seems to be a later, unconnected addition. The Kuwalanamuwa of İMAMKULU and HANYERİ must surely be surely the same actual person claiming the credit for creating both monuments which seem by their placing to mark the beginning and the end of the ascent to the Gezbel pass. Whether he can be identified with the Kuwalanamuwa who left his epigraph on the Sipylos monument can hardly be determined. And while HANYERİ may reveal its locality as the sacred mountain, seat of the god Šarruma, the relevance of the Storm-God of Ḫalab and the role of the nude goddess to the area of İMAMKULU must remain a matter for speculation.

#### *8. Tutḫaliya IV's hunting monuments*<sup>7</sup>

I have argued that certain monuments discovered across the Anatolian plateau represent cultic installations at sites of sacred stag-hunts conducted by Tutḫaliya IV as recorded in his texts KUB 2.1 and related pieces, his «Festival for all the names of the Stag-God» (McMahon 1991: 63-141; Hawkins 2006). The connection between the Cuneiform texts and Hieroglyphic monuments may be provided by the identification of mountains named in the former as 'all the mountains in which his Majesty travels, hunts and shoots' with actual sites of some of the latter. The famous silver stag-rython of the Schimmel Collection, with its scene of the worship of the Stag-God and his consort, whatever its provenance, was surely used in the celebration of this cult, as was also in all probability the bronze bowl from Kınık dedicated by the eunuch Taprami, showing a scene of the stag hunt.

The prime example of such a cultic installation is provided by the EMİRGAZİ altars with the related fragments. Their long and relatively well-preserved text specifically records the Stag-God, his consort Ala, and the sacred mountain Sarpa on which Tutḫaliya has placed their images and stelae. The Stag-God of Mount Sarpa and Ala of Mount Sarpa are among those named in the festival text, and there can be little doubt that this mountain is the Arisama Dağ, at the foot of which the EMİRGAZİ altars and base were found, indicating the existence there of a sacred precinct. A further precinct may well be sought at or near the summit. The text also names another mountain, mount AXE (written with a logogram, reading unknown), which is perhaps the adjacent Karaca Dağ.

A recently discovered stele ALTINYAYLA shows a scene of libation to the god on the stag similar to that of the Schimmel rython. In view of its proximity to the city Kuşaklı-Sarissa, the probability must be that this stele marks the site of a cultic installation of the Stag-God of the city/Mountain Sarissa appearing in the festival texts, and that Mount Sarissa is the local prominent mountain Karatonus Daǧ, at the foot of which Altınyayla is located.

The long-known inscribed block KARAKUYU was placed by Tutḫaliya IV in the centre of the retaining wall of his water installation at the site of that name. The inscription records these mountains

<sup>7</sup> Ehringhaus 2005, 47-50, 80-89 (KÖYLÜTOLU YAYLA; KARAKUYU, ALTINYAYLA). Hawkins 2006.

'hunted' by Tutḫaliya: Sunara, whose Stag-God appears in the festival texts; one perhaps to be read Saluwanta, which also appears in the texts; and (A)sarpa, which may or may not be the same as Mount Sarpa of EMİRGAZİ. These mountains, especially Sunara, and Saluwanta, and (A)sarpa if correctly read and identified, directly connect the monument with the festival and cult, as does the statement that Tutḫaliya 'hunted' them. It may be that Sunara is to be located in the vicinity of Karakuyu, and so also perhaps Saluwanta.

The KARAKUYU inscription also contains three signs, TONITRUS.PURUS and a third which may represent pictographically a corbel vault or postern, the whole perhaps to be understood as the «the Storm-God's sacred vault», as I suggested in my presentation. Afterwards J. Klinger and A. Schachner kindly drew my attention to a proposal to recognize the structure incorporating the KARAKUYU block not as a sluice, as has usually been thought, but as a cultic chamber comparable to Kammer 2 in the sacred pool complex at Boǧazköy (Hüser *apud* Schachner and Wittenberg 2012, 249 n. 20). This proposal would fit very well with my suggested reading, so «the Storm-God's sacred vault» may be recognized as a reference to the Karakuyu structure itself. Further, these observations would combine to support the suggestion of P. Neve to see the Karakuyu monument not as a «profanes Wassereservoir» but as connected with a «Quellkult» (1992: 75).

A comparable monument to Karakuyu is that from Köylütolu Yayla south of the road between Ilgın and Kadınham, where a large block with a 3-line Hieroglyphic inscription was found in connection with another embanked water-retaining pool (new details from Ö. Harmansah's *Yalburt Yaylasi Archaeological Landscape Project*). The block's original form and purpose remains unclear, since the inscription is incomplete on the right side, apparently also the left, implying the loss of inscribed elements on either side, which has reduced lines 1 and 2 to isolated fragments of text, though line 3 seems to preserve a complete final clause. What can be understood of the text shows traces of a 1st person singular royal speaker, presumably Tutḫaliya IV: compare (line 2) «(He) who (is) my grandson's son» (= EMİRGAZİ 1 §4). Line 3, apparently credits the execution of some part of the monument («These engravings (?) which…» to a high official Šauškaruntiya «Prince, chief of the palace servants, army lord», known also from his seals (Herbordt 2005, nos 373-378; see also Hawkins *ibidem*: 272). The inscription KÖYLÜTOLU YAYLA does not preserve much information about itself or its locality beyond an association with a city name, twice recurring, written TA5 -*tara/i-ma* (URBS). Evidence has recently accumulated that the sign TA5 (L. 172) has the value *(a)la* (Hawkins *apud* Herbordt, 2005: 248, 289), which gives the reading of the city name here as Alatarma. The known town of this name is far away in the east on of the Euphrates: a man of Alatarma is one of the oath -takers in the Mita of Paḫḫuwa text, and it is to Alatarma that Tutḫaliya IV himself fled after his defeat to the battle of Niḫriya. So what could Alatarma be doing at Köylütolu Yayla, the other end of the Hittite Empire? Here an etymology of the toponym might serve to explain the recurrence here of a local Alatarma, if we may analyse it as *ala*-, Luw. «high», *tarma*-, Hitt.-Luw. «peg., point», thus «High Peak». Mountains might be named from their local towns, cf. city/mountain Sarissa in the festival texts (above), which indeed list both a Stag-God and consort Ala of Alatarma. So for an Alatarma near Köylütolu Yayla we might seek no further than a local «High Peak». And if the Stag-God and Ala of Alatarma belong to this one rather than its eastern namesake, this would connect the KÖYLÜTOLU YAYLA monument too with the site of a cultic hunt precinct of Tutḫaliya.

One further monument may be included in this category of markers of cultic hunt precincts: the crude little Delihasang stele from a village near Boğazköy on the Büyük Nefesköy road. The middle line of this stele is occupied by the usual great cartouche of Tutḫalya IV, while the top line, of which only the right side survives, preserves only two signs, 'every mountain'. This alerts us to the recurrent phrase in the festival texts, «all the mountains which his Majesty travels, hunts, and shoots». With this in mind I would be reasonably confident in restoring the missing left half of line 1 as (DEUS)CERVUS4 (L.463), «(to) the Stag-God (of the Countryside?)», indeed surviving traces are consistent with the reading CER[VUS4 ].

#### *9. Yalburt*<sup>8</sup>

Distinct from these hunting monuments of Tutḫaliya IV is his YALBURT structure, some 23 km. north-west of Köylütolu, a group of at least twenty inscribed blocks found out of order but restorable

<sup>8</sup> Rossner 1988, n. 5, 63-66; Ehringhaus 2005: 37-47.

as the upper course around three sides of rectangular built pool, fed by a natural spring. The text as preserved is a historical account of Tutḫaliya's campaign against the Lukka lands and contains no reference to the monument or its *raison d'être* and location at the spring, though this may be assumed to have been sacred like most (all?) significant water sources.

#### *10. Karabel and Hatip*<sup>9</sup>

Two very similar contemporary monuments might seem to have a political rather than a religious function: the inscribed rock reliefs of KARABEL and HATIP, western and central Anatolia, showing each a figure with bow and spear and identifying epigraph, the one Tarkasnawa king of Mira, the other Kuruntiya king of Tarḫuntassa. Both appear to have a primary function as frontier markers. The Karabel pass over the Boz Dağları (classical Tmolos) carries a road from Ephesos to Sardis, and in terms of Late Bronze Age geography is best identified as as the frontier between the Arzawa lands Mira (to the south) and the Seḫa River land (the Hermos – Gediz river, to the north). There is nothing obviously sacred about the location of the relief except in so far as a pass or frontier might attract that sanctity. HATIP on the other hand is placed on a rock face with an abundant spring at its foot, a feature very likely sacred which itself might have attracted the relief. That it also stands on the frontier between Hatti and Tarḫuntassa may be inferred from the Bronze Tablet treaty, which shows that Ikkuwaniya (Konya) was in Hatti, while Ussa (probably Karahöyük – Konya) was in the Ḫulaya River land, i.e. Tarḫuntassa. Its frontier description runs from Pedassa and the sacred mountain Ḫuwatnuwanda, apparently along the hills to the west of Konya, then turns east through south Konya plain to the Cilician Gates. HATIP would have stood right in the vicinity of such a frontier.

#### *11. Karadağ – Kĭzĭldağ*<sup>10</sup>

A fine example of a sacred mountain with a peak precinct is the Karadağ, an isolated massif in the south Konya plain, which is linked to the Kĭzĭldağ, a small but steep peak at its foot with a settlement, probably citadel and lower town. The two sites are connected by the presence on both of inscriptions of a certain Hartapu son of Mursili, both bearing the titles «Great King, Hero», while Hartapu additionally awards himself the winged disc, «My Sun». The date of these inscriptions has been much discussed, but their close connections with the Empire period style are strong enough to point to a shortly post-Empire date or even earlier (The incised figure of Hartapu on Kizildağ on the other hand can hardly be earlier than the 8th century BC, so I think that it has to be accepted as a later addition).

The KARADAĞ inscription on its highest peak Mihalĭç is incised on the parapet of a rock-cut corridor, the end of which is blocked by the fallen ruins of a Byzantine church on the summit. For one looking over the top of the parapet there is an uninterrupted view straight down to the Kĭzĭldağ some 13 km distant to the north-west. The inscription reads: «In this precinct the Storm-God of Heaven, the divine Great Mountain (and) all the gods My Sun, Great King Hartapu…» (unread sign, verb 'honoured' or the like required). It is clear that the 'precinct' must have been the mountain-top shrine lying now under the ruined church, which had doubtless been attracted to the site by its on-going sanctity. The «divine Great Mountain» must be the Karadağ itself and could well be the «Great Mountain» attested in Hittite texts (ḪUR.SAG GAL / *RAB*Û), though it is perhaps rash to assume that there was only one mountain so designated in Hittite Anatolia.

A further inscription of Hartapu was found at Burunkaya some 140 km north-east of the Karadağ, beyond Aksaray. It was incised on a huge block fallen upside-down from a crumbling bluff on to the hillside below. The BURUNKAYA inscription reads: «In this precinct Great King Hartapu, beloved (?) of the Storm-God, of Mursili, Great King, Hero [son…]». This indicates the existence of a sacred precinct on the original top of the bluff, though there is nothing very obvious to suggest what natural feature might have attracted the sanctity.

<sup>9</sup> KARABEL: Kohlmeyer 1983, nos. 2-9, 12-28; Taf. 1-8; Rossner 1988, no. 2, 46-52; Hawkins 1988; Ehringhaus 2005: 87-91. HATIP: Ehringhaus 2005: 101-107.

<sup>10</sup> Ehringhaus 2005: 14-33.

#### 12. NA4*ḫekur* SAG.UŠ11

We now return to consider further the two cuneiform Hittite terms the NA4 *ḫekur* and the DKAS-KAL.KUR. This takes us back to Boğazköy-Ḫattusa itself, specifically to its sites known as Nişantepe, Yazılıkaya, and the Südburg NA4*ḫekur* SAG.UŠ. The remarkable tablet, KBo 12.38 has two colums each side, each truncated to an isolated fragments of text by the loss of the tablet's top and bottom: col. I, part of a report of a Hittite king's campaign against Alasiya (Cyprus) with taking of tribute; col. II, a report by Suppiluliuma II of making a statue of his father Tutḫaliya IV, writing his annals on it, and setting it up in an 'eternal *ḫekur*'; then a double ruling and a new text beginning «I am Suppiluliuma (genealogy). [My father] Tutḫaliya […] (breaks off)»; col. III, a report of a naval battle fought by Suppiluliuma off Alasiya, followed by a landing and battle; col. IV, Suppiluliuma builds «this eternal *ḫekur*» for a statue, installs and endows it, then bans its misappropriation or subjection to taxation on pain of divine retribution (end of text).

Otten in his first publication of text inferred that the NA4*ḫekur* with the inscribed statue of Tutḫaliya was Yazılıkaya chamber B, the character of which seemed to fit the description. Güterbock showed that the tablet contained two separate texts: the first narrating an Alasiya campaign [of Tutḫaliya] as part of his annals inscribed on the statue in Hieroglyphic Luwian; the second beginning after the double ruling with «I am Suppiluliuma […]» typical of a Hieroglyphic Luwian but not Cuneiform Hittite text, continuing with Suppiluliuma´s own annals, inscribed on «this eternal NA4*ḫekur*». He thought that this would have been an inscription «*comparable* to Nisantaş and dealing with a building on a mountain peak *comparable* to Nişantepe». In fact I believe that both Otten and Güterbock were correct. Yazılıkaya chamber B should be Tutḫaliya's eternal NA4*ḫekur* containing his inscribed statue. Nişantepe should be the site of this eternal NA4*ḫekur*, built by Suppiluliuma [for himself] and NİŞANTAŞ his own annals inscription.

Investigation on Nişantepe by Neve after Güterbock´s suggestion planned the beddings for building blocks on the summit and revealed the ascent ramp with fragments of a fallen sphinx gate, suggesting the type of building which had stood there, the presumed NA4*ḫekur*12. I myself have worked intensively on NIŞANTAŞ for two seasons, 1992-1993, and have a number of probable readings for line 2-6 beyond the long-recognized line 1, which serve to confirm its character as an annalistic inscription of Suppiluliuma II, without being able to identify anything to tie it to what is preserved on KBo 12.38, cols. III and IV. A 3-D scan on the inscription was made in 2014, which together with my readings formed the subject of a workshop, held in Würzburg University in March 2015. I have prepared an edition of the NIŞANTAŞ inscription for my forthcoming corpus of the Hieroglyphic Luwian inscriptions, vol. III.

#### *13. DKASKAL.KUR*<sup>13</sup>

In conclusion we return to a consideration of the Cuneiform Hittite term DKASKAL.KUR. These are invoked as divine witnesses after mountains, rivers, and springs, and named as geographical fixed points in frontier descriptions. They are thus divinized topographical features connected with water, and have with great plausibility been identified by E. Gordon (1967) as sink holes, places where rivers disappear underground, geological formations common on the Anatolian plateau. It is easy to see how these could be regarded as entrances to the underworld, as is implicit in the literal interpretation of the compound logogram, «divine road of the earth».

Chamber 2 of the Südburg sacred pool complex at Boğazköy excavated by Neve in 1988 incorporates a long and well preserved Hier. Luwian inscription. The text is a historical account of his sacred conquests by Suppiluliuma II, but terminates with a statement describing the structure itself: «Here a DEUS.VIA+TERRA in that year (I) made». Since these three Hieroglyphs correspond exactly sign by sign to DINGIR.KASKAL.KUR, an obvious suggestion was this was an *ad hoc* Hieroglyphic creation to render the Cuneiform concept, thus that the Südburg complex was an artificial DKASKAL.KUR. It remained to explain how this artificial construct could be described by a term understood as a natural feature. The excavator, independent of this epigraphic evidence, interpreted the complex as connecting with an underground water cult, and in this sense it might be designated a DKASKAL.KUR.

<sup>11</sup> Güterbock 1967.

<sup>12</sup> Neve 1992: 64-66.

<sup>13</sup> Otten 1980; Hawkins 1995; Ehringhaus 2014 (İvriz).

#### 8 *JOHN DAVID HAWKINS*

For one attestation of a DKASKAL.KUR an actual identification may be proposed. In the Hatti-Tarḫuntassa frontier description (Ulmi-Tešub Treaty // Bronze Tablet Treaty, §7) we read: «(Coming) from the city Zarwisa, Mount Sarlaimmi, the DKASKAL.KUR, the water of the *ḫinnaru*(*wa*) is his frontier». Mount Sarlaimmi (Luw. «Exalted») is generally agreed to be the Toros / Bolkar Dağ massif, so this DKASKAL.KUR and the *ḫinnaru(wa)*-water (whatever that might mean) might well refer to the İvriz Spring. This is in fact no ordinary spring but a cavern from which seasonally with the melting snow a colossal torrent gushes for a few months but is dry from the autumn to spring. Additionally further down its watercourse, multiple spring well up just before the spur of rock on which the famous İvriz relief is carved. The whole area, particularly the cavern, has a character which we might expect to be designated a DKASKAL.KUR. Evidence of the awe which the cavern still attract is the *çaputlu agaç* (tree with voting scraps of cloth attached to ist branches) growing at the cave mouth.

One attestation of DKASKAL.KUR has been cited against Gordon's interpretation «sink hole», because it indicates that one goes up to a DKASKAL.KUR at a high point: (the king goes from Ḫanḫana to Ḫattena by chariot), «and when the king and the queen arrive up at the DKASKAL.KUR, they proclaim it a look-out post» (Otten 1988: 33 f; text now published as KUB LX 148). But this does not necessarily contradict Gordon's interpretation. If the Südburg is indeed a (artificial) DKASKAL.KUR, one can certainly go up to it and enjoy a good view. And even if the DKASKAL.KUR at Ḫattena was a natural feature, it is probably possible to find a cavern with water high up on a hillside which could attract this designation.

#### *Bibliography*


Hrouda B. 2011, *Sirkeli*, «Reallexikon der Assyriologie und vorderasiatischen Archäologie», 12: 554.


# HITTITE RELIGIOUS LANDSCAPES

*Alfonso Archi*

#### *Abstract*

Hittites had longer periods of time devoted to fulfilling religious duties. The crucial ceremonies of the AN.TAḪ.ŠUM and the *nuntarriyasḫa*-Festivals, concerned yearly renewal in a cosmic dimension. Their geographical setting was more or less within the territory of Hattian Hattus(a). The restyling of the Spring and Autumn festivals at the time of Suppiluliuma and Mursili II extended the celebrations over many days, introducing major Hurrian cults according to the customs of the towns of Halab and Hattarina. Particular aspects of the Hittite festvals, especially their economical, political, and geographical implications in light of written sources of the late Imperial period will be addressed in this article.

\*\*\*

#### *1. Another concept of time*

The rhythm of seasons, assuming the character of religious cycle, receives a metaphysical meaning1 . 'Sacred time' is marked by exact anniversaries and may include elements of social and historical origin: Israelitic festivals commemorate fundamental events of the history of salvation.

The only sacred dimension of time which interfers strongly in contemporay daily life is Ramadan in the Islamic societies, which falls in the 9th month of the traditional Muhammadan year. Christian Lent (which is ten days longer) originally had even stricter prescriptions, but these have generally been abolished. Traditional societies usually had much longer periods of time devoted to fulfilling religious duties. This is also the case of the Hittites. The Festival of Spring, which received the name of AN.TAḪ. ŠUMSAR from a crocus deposited in the temple of certain gods, lasted 35/38 days in the early period of the Empire, and was extended to 40 days in the time of Tutḫaliya IV. Correlated with this was the Festival of Autumn, named «of Haste», *nuntarriyašḫaš*, which also lasted about 40 days under the same king. The king and queen had the duty of celebrating in person most of the rites for all the prescribed days.

The festival of *purulli*, related to the New Year, must have been rather long in the form celebrated at Nerik, whose prescriptions needed 32 tablets (KBo 31.8+KUB 30.42(+) I 5-7; Dardano 2006: 22-23). Another version was celebrated directly by the king at Ḫattusa (Götze 1933a: 188-191, ll. 38-47). The annual *ḫišuwa* festival, addressed to the gods of Kummani, in the version introduced from Kizzuwatna by request of queen Puduḫepa, Ḫattusili III's wife, lasted nine days. There are several other festivals celebrated by the king, the queen, and the (heir) prince (in part preserved in fragmentary state); some of these, however, were probably included in the AN.TAḪ.ŠUM or *nuntarriyašḫaš* festivals already in the early Empire, although this is not explicity stated. Another festival was celebrated in cycles of six years (Götze 1933a: 138-139, ll. 40-41: «I wintered in Ḫattusa, and I celebrated the festivals of the sixth year». Moreover, besides the yearly festivals, the Month festival at Ḫattusa also required the royal couple perform as officiants2 .

<sup>1</sup> See, in general, Berlejung 2003; Rüpke 1996. For some remarks concerning the sacralization of time by the Hittites, see Hutter 2008. 2 The Month festivals are studied by Klinger (1996: 286-614).

#### 12 *ALFONSO ARCHI*

It is not possible to calculate even in an approximate way the number of days required of the king and queen for these religious duties. Already the fact that the Spring festival needed more than one month before the king moved to war (another of his yearly duties), and that so many days had to be devoted to the Autumn festival, when he returned to the capital, gives a clear perception of the burden of the cultual obligations bound to the Hittite kingship, considering also the climatic conditions together with the morphology of most of Anatolia, which rendered military operations difficult in all seasons of the year (Klengel 2010: 180-181). No wonder that when the king was engaged in demanding campaigns in far away Syria, he had to neglect such duties. At the beginning of his *Ten-Years Annals*, Mursili II states: «Because my father was garrisoning in Mittanni, he tarried in garrison; the festivals of the Sun-goddess of Arinna, my lady, were being therefore neglected (i.e. not celebrated)» (Götze 1933b: 20-21, ll. 16-18; *CHD* Š: 59).

#### *2. An administrative landscape: the KI.LAM Festival*

The KI.LAM Festival (a pseudosumerogram for Hittite *ḫilammar*, «gatehouse») took place in Ḫattusa and was addressed to the gods of the capital, including the Sun-goddess of Arinna and the Storm-God of Zippalanda, the two major holy cities. The exemplary edition by I. Singer (1983, 1984a) has to be completed in some parts by a certain number of small fragments published in later years. The festival, which lasted three days, goes back to the Old Kingdom period (several manuscripts are in old and middle Hittite script); a new version was drawn up in the 13th cent. It was addressed exclusively to Hattian gods (listed in Archi 1993: 5) and includes several invocations in Hattian (Groddek 2004)3 . Although this festival is perhaps the best attested in terms of the number of manuscripts, it is not mentioned among those listed in the Instructions for Temple Officials (Süel 1985; Taggar-Cohen 2006: 33-107). This is a good indication, therefore, that it was included in a seasonal festival. In fact, the KI.LAM is mentioned in the 25th, 26th and 27th days of the *nuntarriyašḫaš*, whose rites were celebrated in part in the temple of Halki, the Grain-goddess (Nakamura 2002: 80-81; 127-130)4 . The king sits first outside the gate of the palace to inspect the procession of cult symbols and 'animals of the gods' made of precious metals, introduced by the priest of Inar(a) (later manuscripts: DKAL), the tutelary deity of nature and the wild animals. The royal couple then rides in chariots to the gate of the temple of the goddess Halki. The following ceremony is that of the «administrators», LÚAGRIGMEŠ, of several towns, which takes place in the vicinity: they stay at 'the gate of (their) houses', KÁ É, and offer victuals as breads, beverages and livestock (Singer 1983, 59-63, 157-167). Another station of the royal procession was the gate of the goddess of growth Miyatanzipa (Singer 1984a: 78, rev. V 3'-8'). While those deliveries were presumably consumed by the partecipants to the festival, this ceremonial procedure by the temple of the Grain-goddess clearly symbolies the supply of victuals to the capital by the towns of central regions of the kingdom through the administrators at the head of «storehouses», É NA4KIŠIB, located in those same towns. It is quite uncertain as to whether this administrative organization had at its disposal real 'houses' in Ḫattusa, or whether these were just symbolized in some way on the occasion of the celebration of the festival at Ḫattusa5 .

The preserved names of the towns whose «houses», É, (with their administrators) supplied victuals are:

*a*) 2nd tablet: KBo 10.24 IV 18-V 28; 3rd tablet: KUB 10.1 I 2'-4' (and duplicates; Singer 1984b, 20-22) URU*A-an-ku-wa* (*Ḫa-ni-ik-ku-i-il*: Hattian «he of Ankuwa»)

URU*Ne-na-aš-ša*

URU*Tu-u-wa-nu-wa*

<sup>3</sup> A synopsis of the KI.LAM festival is given by Singer 1983: 58-64; see, further, the descriptions by Haas 1994: 748-771 (the sections concerning Arinna and Zippalanda at pp. 767-771, belong, however, to the AN.TAḪ.ŠUM Festival), and Görke 2008: 51-57. 4 An oracle account (ABoT 14(+) III 8-19; duplicates are KBo 30.22 and KBo 30.23) concerns the provisions for

the KI.LAM Festival celebrated in the temple of Ḫalki, see Singer 1983, 134-135. It was Houwink ten Cate (1988, 191-193 with nt. 53) who has suggested that the KI.LAM festival was performed during the Autumn Festival of *nuntarriyašḫaš*, not excluding, however, the possibility that it may have been included in both the big seasonal Festivals.

<sup>5</sup> On the supply system, in particular in relation to the religious festivals, see Archi 1973a. The AGRIG officials have been studied extensively by Singer 1984b.

URU*Ḫu-u-pí-iš-na* [ URU .....*y*]*a* [ URU.........] URU*Ka-at-ti-la*


Towns whose administrators supply victuals (ration lists):


The AGRIGs usually appear in function of their administrative duties in religious documents (the number of the administrative documents from the archives of Ḫattusa is minimal). VBoT 68 contains in the preserved part of the tablet three lists of AGRIGs who are presented to the king by the herald, who names them by the town to which they belong, adding the Hattic derivational suffix *-il* (e. g.: URU*Zalpuīl* «that from Zalpa») (Götze 1930). The introductory formula of the first list is not preserved; the second and the third ones have respectively: «When the king goes out from Arinna to Ḫattusa the herald calls out the *telipuri* as follows / When the king goes from Ḫattusa to Matilla the herald announces the AGRIGs to the king as follows». The *telipuri* was therefore the administrative 'district' within which each AGRIG was active (Singer 1984b: 118-119). This document undoubtedly preserves a structure of the Hattian administrative organization: not only that of the ancient city-state of Ḫattusa but that of the Hatti region (*Ḫattum*, according to the Assyrian), that is the region inside the bend of the Kızılırmak. In fact, among the cities listed in VBoT 68 there are Zalpa (on the Black Sea), Ḫurma, Durmitta, and several others independent cities of the time of the Assyrian merchants.

#### 14 *ALFONSO ARCHI*

The KI.LAM Festival proves that this administrative system had assumed a religious dimension already at the time of the Hattians, which the Hittites preserved and adapted for the core of their kingdom: in autumn the administrators delivered personally in Ḫattusa the victuals for celebrating the festivals of the ancient deities of the official pantheon. It is meaningful in this respect, that the AGRIGs were presented to the king and the queen in front of the temple of Halki. The name of this goddess is the common term for 'grain' in Hattian (and Hittite): she represents therefore the basic form of nourishment. Significantly, the AGRIG of Arinna does not appear in these documents: it was the duty of all the other cities to provide the victuals for the Sun-goddess of Arinna, the deity at the head of the pantheon, at the great temple of the capital, Ḫattusa, as well as in her sanctuary at Arinna.

The AGRIGs of the KI.LAM festival (whose list is probably not complete) belong not only to the central area (Ankuwa is Alișar; Ḫarḫarna ?), the Upper Land (Karaḫna) and the eastern region (Sukziya, Wattaruwa ?), but also to the Lower Land: Ḫupisna, Nenassa, Tuwanuwa, Zallara (the localization of many other cities remains uncertain, as for Kastuwara, Kattila, Zikkurka, Zinirnuwa). The geographical extension is very large, and the towns of the Lower Land were later provided with a «storehouse», É NA4 KIŠIB, at whose head was an administrator (AGRIG) according to the Edict of Telepinu. Several other AGRIG lists and ration documents in old script are related to religious celebrations (not necesserely the KI.LAM Festival), confirming the religious dimension of the system which provided victuals6 .

The Assyrian document from Kanesh do not depict Hattum as a unified kingdom, but, on the contrary, as a region with several rival cities, Zalpa and Ḫattusa being the two most important kingdoms, at least at the time of Anitta. The Hittites, therefore, extended the originary Hattian administrative system to the regions included in their kingdom, and it is this state of affairs which is reflected in the Hittite version of the KI.LAM Festival.

The Hittite received from the Hattians not only the gods for their official pantheon together with the related cultual ceremonials, but also the belief that the land had to supply its gods with food consigned yearly by those who represented the central administration in each of the major centres. Similarly, all the provinces of Assyria in the 12th and early 11th cent. (and also later) had to feed their god Assur with «regular offerings», *ginā'u* presented in his temple in Assur (Maul 2013).

# *3. A political landscape: the Festival for the Storm-God of Zippalanda (CTH 625) (37th day of the AN.TAḪ. ŠUM Spring festival)*

#### *3.1 The texts*

The tablet KBo 4.13+KUB 10.82, although rather well preserved, lacks about the first 5 / 8 lines in col. I, and the colophon, which makes it difficult to classify it exactly. It belongs to the AN.TAḪ.ŠUM Spring festival because of an offering of the AN.TAḪ.ŠUMSAR plant (together with breads, a stewed dish, beer and wine)7 ; consequently, E. Laroche classified it as CTH 625. V. Haas (1994, 775-781) has suggested that this text could concern the rites of the opening ceremony of the Spring festival, celebrated at Ḫattusa (as 'Day 0'). This is, however, not in agreement with the incipt of this festival (which started, of course, with 'Day 1'): «If the king spends the winter up in Ḫattusa, then, when the king sets out from Ḫattusa, the king and queen go from Ḫattusa to Taḫurpa. But if the king spends the winter anywhere else, in another town [e.g. Mursili II used to spend the winter in Ankuwa (Alișar Höyük)] […] from that town he goes to Taḫurpa», KBo 10.20 I 2-10 (Güterbock 1960: 80, 85)8 . In this second case it is clearly stated that the king did not have to go to Ḫattusa to open the Spring festival, which, having a cosmic value, started with an *inauguration act*: *the entering of the royal couple into the capital*.

KBo 4.13(+) cannot be even the celebration in Taḫurpa («in the *ḫalentu* house (of Taḫurpa) (where) the great assembly (takes place)» (ll. 17-18), or that held «in the *ḫalentu* house» in Ḫattusa (l. 23; 'Day 2'). This is because in KBo 4.13(+) the Storm-God of Hatti, the major male god of Ḫattusa (and of the Hittite pantheon) is never mentioned, The three principal gods of the festival are the Sun-goddess of

<sup>6</sup> For a geographical distribution of the AGRIGs mentioned in the religious documents, see Singer 1984b: 120-127.

<sup>7</sup> Vs. III 18-20: *ta IŠ-TU* NINDA KU7 NINDA*pu-un-ni-ki-it* UTÚL*ḫu-ru-ti-li-it* AN.TAḪ.ŠUMSAR *IŠ-TU* KAŠ GEŠTIN *AŠ-RI*<sup>Ḫ</sup>I.A *ir-ḫa-an-zi* "Then with sweet bread, *p*.-bread, ḫ*.-*stew, AN.TAḪ.ŠUM plant, beer (and) wine they make the rounds of the (Holy) Places («of the temple» (cfr. the parallel passages in KUB 2.8 II 1-21, CTH 617; for KAL of Taurisa, celebrated on the 32nd day).

<sup>8</sup> Schwemer (2004) has discussed in length the ordering of the fragmentary documentation for the beginning of the festival, supporting Haas' thesis.

Arinna, the Storm-God of Zippalanda (in the second place) and the Tutelary-god (DKAL) of Hatti, who are followed by a long list of other deities, as M. Forlanini (2007: 259) has stressed.

The position of the Storm-God of Zippalanda immediately after the Sun-goddess of Arinna (the major deity of the pantheon), in the place of the Storm-God of Hatti, can only mean that this celebration took place in Zippalanda9 . This is in agreement with the outline *F* of the Spring festival. The king's journey was in a southerly direction from Ḫattusa (as Güterbock 1961: 92 has already remarked) according to the colophon of the text concerning the journey of the divine Hunting bag of the Storm-God of Zippalanda, KUB 20.25+10.78 VI 11: *A-NA* KASKAL IM.U19.LU *pa-iz-zi* (cfr. Popko 1994: 308-309).10 On the 36th day, the king arrived in Zippalanda from Hurranassa: cult action «at the *eya*-tree of Zippalanda».

The (following) morning (i.e. on the 37th day) the king celebrates [the festival of the AN.TAḪ.ŠUM plant] to the Storm-God of Zippalanda in his temple; (to) the Sun-goddss of Arinna and the Storm-God of Hatti at the *eya*-tree [of Zippalanda]. The (following) morning (i.e. on the 38th day), on the Mount Daha: festival of the invocation (*ḫalziyauwas*) (Rs. 29-30; Howink ten Cate 2003: 210).

The deities listed in KBo 4.13(+) (excluding those of the long god-list in Vs. I) are the same as KBo 19.128, as H. Otten has noted (1971: 27-48), whose colophon reads: «Second tablet; complete. Concerning the Great House (É*TIM* GAL): of the Spring (festival). (With this one) the 2nd day is ended. According to the wooden tablet (GIŠḪUR *ḫandan*). (This is) the tablet (*ṬUP-PU*) of Ḫattusa. Written by Arma-ziti, the scribe, under (the supervision of) Anuwanza, the LÚSAG».

This tablet, in clay and written for the archives of the capital, is to be dated, therefore, to the time of Tutḫaliya IV, as is also KBo 4.13(+), since both share several similar elements. The god-lists in KBo 19.128 I 39 ff., II 19 ff. and KBo 4.13(+) II 9 ff., III 28 ff., IV 14 ff. are the same, all opened by the Sungoddess of Arinna with her daughter Mezzulla, the Storm-God of Zippalanda and KAL (Inara) of Hatti. The Storm-God of the Great House appears in KBo 19.128 II 7, 44 as well as in KBo 4.13(+) II 18, IV 1, 21 (Otten 1971: 29). The long list of gods which opens KBo 4.13(+) (col. I) must have been in the missing first tablet of the other manuscript.

One has to infer that: 1) both texts concern the Spring festival celebrated in the 'Great House', a building which the king reached by a light chariot (GIŠ*ḫuluganni-*) from the rest-house (É*ḫalentuwa*) 11; 2) the festival was celebrated in Zippalanda for the Storm-God of this town, with whom the Sun-goddess of Arinna was associated for that particular occasion. According to its colophon, KBo 19.128 was the celebration of 'the 2nd day'. This difficulty may be overcome if one considers that Outline *F* says that the king (surely with the queen) arrived in Zippalanda the previous day (as said above): «The ( following) morning the king celebrates the Sun-goddess of Arinna in Hurranassa (where they had arrived the previous day). (Then they go) to the *eya*-tree of Zippalanda. 36th day» (Howink ten Cate 2003: 210). The 'Great House' (É*TIM* GAL) should be, therefore, the temple of Zippalanda, possibly renewed by Tutḫaliya IV, for which the neologism É*TIM* GAL was used. Similarly, in a text written at the time of Tuhaliya IV (see here below), É*TIM* is the writing used for a temple in Ankuwa, KUB 11.27 (+) KUB 41.55 I 15': DU É*TIM* URU*An-ku-wa* "the Storm-God of the house (i.e. of the temple) of Ankuwa" (Archi 2006: 157).

The Autumn and Spring festivals of the Storm-God of Zippalanda, celebrated at Zippalanda and at Mount Daha has along textual history, going back to the Old Kingdom, and collected by Popko (1994: 92-218). As the celebration for Kattaḫa («the Queen») in Ankuwa (KUB 11.27(+)41.55 // ABoT 24) was totally changed at time of Tutḫaliya IV (see here below), it is quite possible this was also the case of the festival for the second Storm-God of the pantheon.

#### *3.2 The sequence of the gods in the opening list and the growth of the kingdom*

M. Forlanini (2007) has noted with insight that part of a long list of deities in the 48 preserved lines in col. I of KBo 4.13(+) follows a geo-political order given by the historical growth of the Hittite kingdom.

<sup>9</sup> Popko (1994: 34) remarks that «in seinem ziplandäischen Tempel wird er (the Storm-God) samt der Sonnengöttin der Erde und dem himmlischen Wettergott mit Opfern versehen».

<sup>10</sup> The following leg of the Hunting bag was Ankuwa, KBo 30.155 VI 2-5: «While the Hunting bag goes by the southern road (KASKAL IM.U19.LU *pa-iz-zi*), and while it reaches Ankuwa from Zippalanda […]» (Popko 1994: 310-311). The king travelled by a chariot (GIŠGIGIR); he reached Mount Daha from Zippalanda by a light chariot ( GIŠ*ḫuluganni-*), KUB 20.96 XX 19-21, Popko 1994: 192-193.

<sup>11</sup> The É*ḫalentuwa* of Zippalanda is quoted e.g. in VSNF 12.215 II 7, see Popko 1994: 238-239.

#### 16 *ALFONSO ARCHI*

The missing first 5/8 lines at the top of the column should have had a kind of introduction to the festival. The following 16 lines have an unparalleled list of gods opened by [Heaven] and Earth; the Fate-goddesses; the Hattic dyad Tašammat and Tašimmet; three tutelary spirits: of the Forehead, of Speaking, of Sin (Hantassa, Lalassa, Wasdulassa, all presenting a Luwian suffix). There then follow three aspects of the Storm-God assuring three kinds of favourable (*aššauwaš*) conditions; the courts («the Seven», DVII.VII*BI*) of the Sun-goddess, the Storm-God and the Tutelary god; several other minor deities. This splitting of the divine powers is typical of the period of Tutḫaliya IV (cfr. KUB 2.1)12.

For our understanding of the religious climate of the time of Tutḫaliya IV, and how different constituents coexisted and were taken in account in reorganizing the cults in different centres, it is useful to compare this text concerning Zippalanda, which preserves the Hattian-Hittite tradition both in the rites of 'drinking' and breaking bread to the gods as well as in the pantheon listed in coll. II-IV13, with the celebration for the temple of Kattaha («the Queen») in Ankuwa, KUB 11.27(+)41.55 // ABoT 24), on the 38th / 39th14 and last day of the AN.TAḪ.ŠUM festival (also introduced by Tutḫaliya IV), which shows instead an extremely marked Hurrian influence (Archi 2006: 156-158).

The list in ll. 17-48 follows a quite different principle15. At its start there is the Storm-God of Heaven (DU *ŠA-ME-E*), followed by the Storm-God of Zippalanda. Since this god «occurs alone, i.e. without the gods of Hatti and Arinna, his presence here could reveal something about the shrine, where our ritual was intended to be performed» (therefore: Zippalanda!) (Forlanini 2007: 263). Wahisi, who follows, was a god of the towns of Kartabaḫa, Salma(/Salamba) and Kadaba, which sent their representatives «to the ritual of Zippalanda according to an old Hittite tablet of the KI.LAM festival»16. From this point on, the towns mentioned in relation with the deities are considered. Hurma (known also from the documentation of *karum* Kaneš Ib) was connected with Zippalanda; Sallaḫasuwa had been under the direct administration of Kaneš for some time.

The second section of the list (ll. 20-33) – from Hatti to the town of Uḫḫiwa – refers to Hatti proper («in its ethnic and political meaning»): the land inside the bend of the Halys. The sequence of the towns is almost the same as that in the Old Hittite AGRIG-list VBoT 68 III 1-9 (Götze 1930: 20-23).

The third section (ll. 34-39) concerns the «land of Kaneš», a denomination «still used by Ḫattusili III to define the large geographical area south of the upper Halys». Many of these towns are mentioned in the Old Assyrian Kaneš documents. The Legend of Zalpa «shows that the rivalry between a northern (Hattian) and a southern (Nesite) power was felt by the local population as the result of a long history». This political and cultural distinction was still felt at the end of the Empire period, although the capital of the Hittites (who called their language «the language of Nesa/Kaneš», *nesili*) had been located in the heart of Hatti for about four centuries.

The fourth section (ll. 40-43) «defines the region to the north-west of the Salt-Lake. […] (It) corresponds to the borderland between Hatti and Arzawa, conquered by Hatti for the first time when Sallatiwara was still an important center», actually one of the major ones during the *karum* Ib period (Barjamovic 2011: 350-357). Among the other cities, Lalanda and Ulma are mentioned.

The fifth section (ll. 44-45) includes Ḫassuwa and Tawanaga «east of Maraș», Hatra and Sinuwa «near Malatya». These were «the first Hittite conquests towards the east».

The sixth section (ll. 46-48) concerns «four countries at the periphery of the kingdom». The first is that of Halpa, conquered by Mursili I and later by Suppiluliuna I. There then follow Pala (corresponding to Paphlagonia); and Parsḫunta (Purusḫanda), also the name of a country, the major state west to Kaneš (in the region of Konya), according to the Anitta text. Ussa, the last town, hosted a *wabartum* of the Assyrian merchants; it is likely to have been independent from Purushanda and located further west of it, probably in the environs (or to the north) of Konya.

For the origin of this list of gods in the celebration for the Storm-God of Zippalanda, Forlanini notes that the town of Kappurnanta (I 27) is «unknown to other Hittite texts but frequently mentioned, as Kapurnat, in the Old Assyrian documents». Moreover, URU*U*-*ús*(UŠ)-*sa* (I 48) is an anomalous writing found also in an Akkadian text, KBo 19.95 Vs. 4, and in a single passage in the Palace

<sup>12</sup> This section of the text is discussed in detail by Haas 1994: 776.

<sup>14</sup> 38th day according Outline A; 39 in F.

<sup>15</sup> For a transliteration, and comment, see Forlanini 2007. I follow, from here on, Forlanini's anlaysis, where textual references and literature are given.

<sup>16</sup> KBo 16.71+ I 59, ; KBo 20.3 II 1-12; Popko 1994, 102-103, 108-111. For passages which mention these towns in relation with the cult of the Storm-God of Zippalanda, see Popko 1994: 327.

<sup>13</sup> See the tabulation in Otten 1971: 29.

Chronicle (both referring to the Old Kingdom), while all the other occurences of this name use signs not of the *sin* but of the *shin* series. For these two elements, and other names of towns mentioned above, the list «should go back to a very ancient, independent document, written in Akkadian» (Forlanini 2007: 262-263)17.

#### *4. A geographical landscape: Muwatalli Prayer (CTH 381)*

The very long list in the Muwatalli's prayer KUB 6.45(+) // 6.46, is mostly geographically ordered; it includes the deities of Hatti (the Halys bend); the Upper Land; (Pala-Tumanna, beyond the northern course of the Halys, is missing); Išuwa (II 64-65); Kizzuwatna (represented by its two foremost cult centres, Kumanni and Lawazantiya, but not by other centres in the Cilician plain); the Lower Land (Singer 1996: 172-177), while the first list of the invocation (I 10-18) has to give a theological solution to the Hurrianiziation of the Storm-God, and, therefore, places Ḫebat, Teššub's spouse, among the first deities.

This prayer defines therefore the geography of the kingdom of Hatti from the religious point of view, ignoring the vassal states, although some gods had been introduced from those countries, as the important Teššub of Ḫalab18.

#### *5. The creation of a sacred landscape: the major seasonal festivals*

Some god-lists reflect a theological speculation or a geographical order (section 4). Other lists in festivals address gods with their major cult places according to geo-political principles determined by historical developments (section 3). In the case of the KI.LAM festival, landscape was determined by a social-administrative order (section 2).

The two most demanding festivals of the imperial period: the Spring and Autumn Festivals, which lasted about 38 days each, combined some basic cult elements which went back to the Hattian period, requering a quite restricted landscape, together with an imperial theology. These two festivals were the pivotal events of the religious year. Tutḫaliya IV addresses the Sun-goddess of Arinna in the following terms: «I shall [confess] my sin [before you] and never again [shall I omit] the festivals. I will not again interchange (*namma waḫnummi*) the spring and [autumn festivals]. [The festivals of spring] I shall perform only in the spring, [and the festival of] autumn I shall perform only in the autumn» (KBo 12.58 + 13.162 obv. 7-9; Singer 2002: 108).

#### *5.1 The* AN.TAḪ.ŠUMSAR *Festival*

The festival prescribes that in spring the king must enter Ḫattusa, as he has to take possession of the capital of the state for the inauguration of the new yearly cycle.19 Therefore, «if the king spends the winter up in Ḫattusa, then […] the king and queen go from Ḫattusa to Taḫurpa», where the «great assembly», *šalli ašeššar*, takes place (1st day). Taḫurpa had no particular religious relevance but functioned simply as the starting point of the festival20. Moving from Taḫurpa, the king reaches Tippuwa, which was in sight of Ḫattusa, towards which the king bows. Wine is offered to a stele (*ḫuwaši*) (of the Storm-God?), which the king reaches by chariot whose reins are taken by a man of the bodyguard, in a ritual race (KUB 10.18 I 6-21 // 17 I 2'-15'). The king enters Ḫattusa and descends from his light chariot at the stele of the Sun-goddess (of Arinna) (KUB 10.17 I 26'-28'). Dispositions for receiving the Hunting bag (KUŠ*kurša*) of god Zitḫariya in the temple of the goddess (KUB 10.18 II 1-12 // 10.17 II 2'-8') (2nd day).

<sup>17</sup> Forlanini (2007: 274) suggests that part of this list may go back to a tablet coming from Hurma (a town mentioned in I 34).

<sup>18</sup> Some aspects concerning the introduction of foreign gods by the Hittites have been discussed by Schwemer 2008.

<sup>19</sup> The Outline of the festival was first treated by Güterbock 1960; presented again in Güterbock 1964: 62-68, and reprinted in Güterbock 1977. A transcription which considers all the duplicates is in Alp 1983: 136-149. A detailed description of the festival is in Haas 1994: 772-826.

<sup>20</sup> Notice, however, that according to KUB 25.14 the Sun-goddess of Arinna receives offerings there; in particular, the queen offers a lamb and breads to each of the Sun-goddesses of five previous queens (obv. 23'-32', 41'-49').

#### 18 *ALFONSO ARCHI*

The next four days concern the journey of Zitḫariya's Hunting bag. It arrives from the holy city of Arinna (Alaca Höyük) (3rd day), and the next day it goes to Tawiniya (Tonea?)21, east to Ḫattusa. Then (5th day), it goes to Ḫiyasna; on the same day, in Ḫattusa, the king goes to the roof (presumably in order to look for the new moon; Howink ten Cate 1992: 97-98), while the pithos (DUG*ḫaršiyalli*) of the Storm-God of Zippalanda is brought there. On the 6th day the Hunting bag arrives at Ḫattusa and is taken to the temple of the Grain-Goddess, Ḫalki. The priests of Arinna take part in the celebration with bread offerings, as well as a lamb from Kasaya. During the year, Zitḫariya's Hunting bag was kept in a temple in Ḫattusa. It is evident, therefore, that – as the king – also this tutelary god of kingship had to inaugurate the new year cycle by arriving from outside.

It is on that day that the pithos of the Storm-God of Zippalanda is opened by a great assembly in the palace of the intendants (*abu bīti*). It is important for our understanding of this ceremony to note that it takes place in the residence of the officials in charge of providing victuals. The pithos of the Storm-God of Hatti is opened on the 12th day.

The central performances required by the spring and autumn festivals were the filling (*šuḫḫa*-) of the storage vessel or pithos (*ḫaršiyalli*) with grain in autumn and its «opening» (*kinu*-) in spring. These were the basic cultic actions of the two festivals, because «avec cette cérémonie, durant laquelle on transformait en pain la céréale de l'année précédente, le nouveau produit, à peine gérme, était lié à l'ancien unissant ainsi le cycle agricole d'une année à l'autre, et favorissant ainsi la croissance de la nouvelle récolte» (Archi 1973: 15).

The ceremony of the pithos and its grain, attested by numerous local communities in many cult inventories (CTH 501-530), was a common pattern of the central Anatolian cults, in use since immemorial time, being a practice which in some cases had to be restored (not introduced for the first time) in the second half of the thirteenth century. On that occasion, the cult of a few, and not local gods was introduced. IBoT 2.131 obv. 15'-20':

The vineyards for the god Pirwa are neglected. (There are) 2 *ḫaršiyalli*-vessels: 1 *ḫaršiyalli*-vessel of Pirwa and 1 *ḫaršiyalli*-vessel of [the god Hasgala(?)]. And when Urhi-Teššub reestablished (the cult of) Pirwa, he spoke (as follows): 'As long as they rebuild the vineyards, let the wine be provided by the temple!', and from that day on the *ḫarsiyalli*-vessel of Hasgala is (there). But they do not (longer) provide it (lastly treated by Cammarosano 2012: 11).

These cult inventories were not related to a religious reform, as has often been believed, but to a reorganisation of the cults22.

#### *5.1.1 The core of the* AN.TAḪ.ŠUMSAR *Festival*

The crucial ceremonies of the AN.TAḪ.ŠUM Festival, which went back without any doubt to the Hattian period, concerned yearly renewal in a cosmic dimension. They were:

1) the opening of the pithos of the two main Storm-Gods: those who determined the agricultural cycle through the alternation of the seasons. In the case of the pithos of the Storm-God of Hatti, it was full with *šepitt*- grain (KBo 10.20 II 16) (the choice of this kind of grain is probably a sign of archaism). In a later version of the festival, VS NF 12.1 rev. 10'-14', the pithos ceremony is related also to the Storm-God of Halab (whose cult was introduced in Ḫattusa by Suppiluliyuma I) and that of [the Storm-God (?) of ...-*š*]*andaya* (Houwink ten Cate 2003: 209, 213-214)23. In autumn, during the *nuntarriyašḫaš* Festival, the Storm-God of Zippalanda was venerated at Ḫattusa on the 9th day (instead of the 6th day), while on the 10th day Zitḫariya was venerated in the temple of the Hunt-

<sup>23</sup> Ḫattusili III, establishing the cult of Ishtar of Samuḫa in the property which belonged to Arma-tarhunta, had a stele representing the goddess erected in each town and village, and provided it with a pithos for the seasonal festivals of renewal, IV 72-73: *humanti*(*a*)-*pat* EGIR-*an* NA4ZI.KIN *tittanuškanzi* DUG*ḫaršiali-ya-kan išḫuiškanzi*) (Otten 1981: 28-29).

<sup>21</sup> This equation, together with the identification with Eski Yapar has been suggested by Güterbock 1961: 87. Forlanini (2008: 163-164) prefers instead a location to the south-west of Ḫattusa. Tawiniya cannot be to the south of Ḫattusa, because it was reached in one day from Arinna (Alaca Höyük). For the Old Assyirian sources concerning this town, provided with a *kārum*, see Barjamovic 2011: 297-305.

<sup>22</sup> This was the thesis by Götze 1933: 159-160 = 1957: 169. He has been followed by Archi 2006: 150-153; and now by Cammarosano 2012, who gives a detailed account of the interpretations by Ch. Carter, E. Laroche, Ph. Howink ten Cate, and J. Hazenbos, all in favour (to differing degrees) of a reform by Tutḫaliya IV. Cammarosano believes, on the contrary, that the majority of these inventories are not to be attributed to that king.

ing bag (*INA* É D*Kuršaš*) (Howink ten Cate 1992: 98; see the *Übersichtstafel* 4, KBo 14.76 obv. 5'-11'; Nakamura 2002: 61-62);


#### *5.1.2 Zitḫariya's Hunting bag*

Zitḫariya takes his name from the town of Zitḫara, one of those places close to Ḫattusa where the king could choose to spend the winter (del Monte, Tischler 1978: 513-514). The god lists of the political treaties list him usually immediately after the Tutelary-god (DKAL) of Hatti and before the Tutelarygod of the town of Karahna (see the treaties with Ḫukkana, Manapa-Taḫhunta, and those with the Syrian kings Tette, Duppi-Teššup, Niqmepa, in the tabulation by del Monte 1986: 101). «Tutelary-god of Hatti» was just a general concept. Zitḫariya was instead the ancient tutelary-god having his origin in the Ḫattusa territory, as can be deduced from the AN.TAḪ.ŠUM Festival. In(n)ar(a) (the Hittite reading of KAL; Luwian Kurunta, later Runta) was the lord of wild animals and the hunt, therefore the tutelary-god of uncontaminated nature. For this reason he could also be represented in the form of a Hunting bag made from the fleece of several sheep (Güterbock 1989; Popko 1978: 108-115, especially 109). The god lists, having the aim of including every divine expression (*siue deus siue dea*), add DKAL KUŠ*kuršaš* to D*Zitḫariya*. It seems, however, probable that it was Zitḫariya (i.e. «Inar of the town of Zithara», URU*Zi-it-ḫa-ra-aš* <sup>D</sup>*I-na-ar*, KBo 52.20 Vs. 3) who was the In(n)ar(a) of the Hunting bag par excellence25.

The symbol of Zitḫariya and that of the Tutelary-god of Ḫatenzuwa (a town not far from Nerik), both in the form of a Hunting bag, were kept in Ḫattusa in the «Temple of the Hunting bag» (É KUŠ*kuršaš*), and had to be renewed roughly every nine years, the old bags being sent to a temple outside the capital:

<sup>24</sup> The restoration in the Outline, KBo 10.20 II 11-12: *I-NA* <sup>È</sup>*ḫé-eš-ti*[-*i*] MU.KAM-*a*]*n* was suggested by Güterbock, on the basis of the colophon in IBoT 2.1 VI 8'-10': *mān* MU-*an INA* <sup>È</sup>*ḫeštā ḫamešḫi INA* EZEN4 AN.TAḪ. ŠUMSAR [*pē*]*danzi*, and provided with a wise comment (Güterbock 1964: 67). This text is a fragment concerning the

prescriptions for that day (CTH 609). For the celebrations of the Year, see also Haas 1994: 792-793; Torri 1999: 21-29. 25 According to the invocation KBo 52.20 Vs. 3, Inar of the town of Zitḫara was «brother [of the Storm-God of Hatti(?)]», see Wilhelm 2010: 270-271.

#### 20 *ALFONSO ARCHI*

[When] they renew the [two] Hunting bags [of] Zitḫariya and the Tutelary-god of Hatenzuwa, whether in the ninth year, or whenever they renew them, the time is not prescribed. When they bring the two new Hunting bags of Zitḫariya and the Tutelary-god of Hatenzuwa, they take them into the Temple of the Hunting bag. The place is prescribed: just below the place of the god(-image) the pegs are driven in: they hang them there (KUB 55.43 I 1-10)26.

An oracle inquiry concerning the return of the king from a military campaign and the delivery of Zitḫariya in his temple (referring certainly to the second day of the *nuntarriyasḫaš* Festival), shows that this god (certainly in form of a KUŠ*kuršaš*) accompanied the king on his military campaigns as his tutelary god in war: «When His Majesty (re)turns from the (military) campaign and when they release Zitḫariya to his temple, with respect to the festival which (on that occasion) will be performed for him, 10 bucks […]» (ABoT 14+ V 12'-16'; Howink ten Cate 1988: 180-181). Zitḫariya had therefore the function of protecting the king in war, as well as in his daily life: even the long list of Hurrian gods in the festival for Ištar of Samuha, as it was fixed for Mursili II, includes exceptionally a section with 'tutelary-gods' (DKAL) (KUB 27.1 I 64-67), opened by Zitḫariya of the king and Zitḫarya of the queen» (Wegner 1995: 34, 38, l. 64).

The Hunting bag, made from fleeces, symbolizes prosperity in the Telepinu Myth of Hattian origin, being full of «sheep fat, animal fecundity, wine, cattle and sheep, longevity and progeny». This symbolism, together with the idea of protection in war, goes back to the Hattian culture: Hattian gods, such as Kappariyamu and Kantipuitti, were also represented by a KUŠ*kuršaš* (Popko 1978: 112; Haas 1994: 450). How may these two aspects (war and realm of nature) be reconciled? The noble art of hunting was an activity characterizing kingship (Hawkins 2006). It needed strength and courage; moreover, it involved direct contact with the mistery of uncontaminated lands. For this reason the king and princes were preferably portrayed with bow and spear27.

#### *5.1.3 The journey of Zitḫariya's Hunting bag and that of the royal couple*

Similarly to the king, who had to enter Ḫattusa to start the celebrations of the AN.TAḪ.ŠUM Festival, Zitḫariya, the tutelary god of kingship, also had to inaugurate the festival coming from outside. In form of a Hunting bag, he starts his journey at Arinna, arriving at Ḫattusa in four days, touching upon these towns (as noted above):

*king's and queen's arrival*: (1) to Taḫurpa, great assembly (no rites in Katapa, where the Autumn Festival begins) [1st day] – (2) to Ḫattusa through mount Tippuwa [2nd day] *Zitḫariya's journey*: (1) from Arinna (to Katapa?) [3rd day] – (2) to Tawiniya [4th day] – (3) from Tawiniya to Ḫiyasna [5th day] – (4) (to Ḫattusa) [6th day]

Also in the *nuntarriyašḫaš* Festival, the royal couple and the god Zitḫariya had to enter Ḫattusa from outside (*Übersichtstafel* 1, to be dated to Tutḫaliya IV) (Nakamura 2002: 15-33, 78-139):

*king's and queen's arrival* (1st journey): (1) at Katapa, great assembly [1st day] – (2) (at Katapa for the arrival of Zitḫariya) [2nd day] – (3) (at Katapa) celebration of the Valiant (NIR.GÁL) Storm-God [3rd day] – (4) to Taḫurpa through Ḫisurla (by the river, the king goes through two burning heaps of wood), great assembly [4th day] – (5) the king to Arinna (rites for the Sun-goddess)28; the queen at Taḫurpa: celebration for the Sun-goddess of Arinna and Mezzulla [5th day] – (6) the king and the queen(?) go to Ḫattusa through Tatisga [6th day]29.

*Zitḫariya's journey*: (1) the god enters «the house of the grandfather» (É *ḫuḫḫaš*), that is of Mursili II, at Katapa30; the (heir) prince is also present [2nd day] – (2) the god goes to Hakura(/Hamara), then to Tatasuna [3rd day] – (3) (arrival at Ḫattusa on the 4th day?) (the king enters the *ḫešta*-house; celebration in the temple of the tutelary god, É DKAL) celebration for Zitḫariya in the Temple of the Hunting bag, É KUŠ*kuršaš* [10th day].

<sup>26</sup> The text, edited partially by Otten 1959, has been reedited by McMahon 1991: 143-188, with some pertaining fragments. On the KUŠ*kuršaš* gods, see, further, Güterbock 1964: 67-68; Haas 1994: 454-456.

<sup>27</sup> Several drawings of Hittite monumental representations are collected by Seeher 2009.

<sup>28</sup> See the day festival concerning Tutḫaliya IV in Nakamura 2002, 172-189.

<sup>29</sup> On the *Übersichtstafel* 6, KUB 51.15 (Nakamura 2002: 68-72), and the problem in reconstructing a general Outline of the festival, see Howink ten Cate 1988: 169-171.

<sup>30</sup> DUTU*ŠI* <sup>I</sup> *Mur-ši*-DINGIR*LIM* is mentioned directly in the colophon 1.A.

The AN.TAḪ.ŠUM Festival foresees that the royal couple leaves Ḫattusa only on the last days:

Festival for Mount Tapala; back to Ḫattusa [28th day; Outline G: 30th day; day festival: CTH 593]; (1) at Haitta (at the foot of Mount Piskurunuwa): celebration for the Tutelary-god of the field; offerings are prepared on Mount Piskurunuwa [34th day] – (2) king and queen go up to the Mount Piskurunuwa (the Akca Dağ, 1689m., south to Ḫattusa?): animal offering to the Heptad of Mount Piskurunuwa they sleep at Ḫarranassa [35th day] – (3) celebration for the Sun-goddess of Arinna and [the Storm-God of Hatti?] at the *eyan*-tree of Zippalanda at Ḫarranassa; the majordomo goes to Zippalanda [36th day] – (4) the royal couple is in Zippalanda (Ușaklı Höyük): celebration of the AN.TAḪ.ŠUM Festival, for the Sun-goddess of Arinna and the Storm-God of Hatti at the *eyan*-tree of Zippalanda [37th day] – (5) celebration at Mount Daha (Kerkenes Dağ, ca. 1500m.; day festival: CTH 63531); to Ankuwa (Alișar Höyük), great assembly [38th day] – (6) celebration of the Rain Festival [39th day; day festival: KUB 59.1]

The *nuntarriyašḫaš* Festival foresees, instead, another two journeys for the royal couple32:

(2nd journey) (1) from Ḫattusa through the Zippalanda Gate (King's Gate, Yerkapı or the gate in the lower city leading up to Yazılıkaya and then to Yozgat) to Ḫarranassa (offerings are provided from the 'palaces' of three towns of the Lower Land: Nenassa, Tuwanuwa and Hupisna, as in the KI.LAM festival) [12th day] – (2) Zippalanda (Ușaklı Höyük), celebration of Zippalanda Festival [13th day] – (3) back to Katapa: celebration for the Storm-God of Nerik, great assembly [14th day] – Taḫurpa: great assembly [15th day] – towards Ḫattusa on the road of Tippuwa; at Nirhanta, by the river the king goes through two burning heaps of wood; Ḫattusa [16th day].

The itinerary Ḫarranassa – Zippalanda was prescribed also for the journey of the last days in AN.TAḪ.ŠUM Festival (34th-37th days), including, however, a celebration for Mount Piskurunuwa near the first town. The return journey of the *nuntarriyašḫaš* Festival: Katapa – Taḫurpa – river – Ḫattusa, follows the same route (in the opposite direction) as the outward journey. It is to be noted that at Katapa there was an offering to the Storm-God of Nerik, probably introduced by Tutḫaliya IV.

The Überichtstafel 3 III 21'-23' of the *nuntarriyašḫaš* Festival has a third journey of the king: to Tawiniya, for the 20th day, equivalent to the 31th day of the long list (Nakamura 2002: 51 and 81).

#### *5.1.4 The imperial dimension of the religious landscape of the* AN.TAḪ.ŠUM

The gods to whom the day celebrations were devoted belong mostly to the Hattian tradition. Some gods were introduced later, enlarging this list in relation with political events and the expansion of the kingdom, so that other regions and traditions were represented, particularly as regards Hurrian cults.

21st day (Outline G.: 27th day): in the temple of the Storm-God of Halab, whose cult was introduced in Ḫattusa at least by Mursili II, or, more probably, by his father Suppiluliuma I. Day festival: CTH 698 (Archi 2006: 155-156, 158-160). The Storm-God of Ḫalab is celebrated also on the 28th day of the *nuntarriyašḫaš* Festival.

22nd-26th days: Ištar of Ḫattarina. 24th day: D[URAŠ/NINURTA], and in Outline G: Ištar of Nineveh. Ištar of Ḫattarina is attested from the early Empire period onwards (Houwink ten Cate 2003: 213, 215). In the *nuntarriyašḫaš* Festival, the celebration for URAŠ falls on the 30th day (Nakamura 2002: 135).

27th day: Ninatta and Kulitta, Ištar's servants.

29th day: in the temple of the god Ea. The Hurrian pantheon related to Ea in this temple is listed

<sup>17</sup>th day: [the Storm-God *piḫaššaš*]*ši*, the Sun-goddess of Arinna is followed by the Storm-God of Samuḫa, a town promoted as religious centre already by Tutḫaliya II/III.

<sup>18</sup>th day: Storm-God *piḫaššašši* (whose cult could was introduced from the Lower Land probably by Muwatalli II, if not later), and the Sun-Goddess of Arinna (DUTU URU*Arinna*). His cult was here Hurrianized, because although these two names appear in the colophon of the day festival, KUB 11.13 VI 8'-12' (CTH 613), in IV 10'-23' the Storm-God *piḫaššašši* is followed by Ḫebat, who receives an invocation in Hurrian (Archi 2006: 155).

<sup>31</sup> A Middle Hittite manuscript with the festival for Mount Daha is KUB 34.118 + KBo 20.58, see Popko 1994: 252-256.

<sup>32</sup> A tabulation of the journeys in the *nuntarriyašḫaš* Festival is given by Nakamura 2002: 438-439.

#### 22 *ALFONSO ARCHI*

in the day festival (Popko, Taracha 1988). Outline G for the 29th day has the Moon-god SIN, the Sun-god UTU and Dagan (written: D*Ḫal-ki*).

The most impressive evidence of the late Hurrianization of some celebrations in the AN.TAḪ. ŠUM Festival is the concluding day festival (38th day) according its late manuscripts (CTH 620) (Archi 2006: 156-158).

#### *6. The dimension of the sacred landscape of the Spring and Autumn Festivals*

Geographically, both seasonal festivals took place in an area whose outer limits were within three days' walk from Ḫattusa.

Arinna is identified with Alaca Höyük, 40 km to the north by modern vehicular road. Tawiniya, reached in one day from both Arinna and Ḫattusa (the Tawiniya gate of which must have been in the lower city, facing north), should be located somewhere to the west33.

From Taḫurpa one could also reach both Ḫattusa (the road ran along the foot of mount Tippuwa) and Arinna in one day. Katapa, the starting point of the *nuntarriyašḫaš* Festival, was also one day from Taḫurpa. Katapa lay surely to the east, in the direction of Sapinuwa and the Kaska territory, which also determines the position of Taḫurpa (Forlanini 2008: 151-153; Gavaz 2012 and 2014).

The king's journey concluding the AN.TAḪ.ŠUM Festival was towards the south, if we accept the identifications of Zippalanda with Ușakli Höyük34, and Ankuwa with Alișar Höyük. The royal train, going through the Zippalanda gate, reached Ḫaitta, at the foot of Mount Piskurunuwa, the most important mountain of the festival together with Mount Daha, near Zippalanda. The second day was devoted to the festival of Piskurunuwa, which was followed by a stop for the night in Ḫarranassa. Third day celebrations in Ḫarranassa. Fourth day: arrival at Zippalanda. For the *nuntarriyašḫaš* Festival, the royal train needed only two day's march, not 3. From Kerkenes Dağ (ca 1500 m), supposed to be Daha, it is possible to see the Akca Dağ (1689 m), perhaps Mount Piskurunuwa. G. Summers (2014: 42-43) writes:

Using nothing but a GPS which indicated the direction of our final destination and our own assessment of the topography we walked from Şahmuratlı Köy at the foot of Kerkenes via Kuşaklı Höyük. In two long days 73 kilometres were covered without difficulty. Had we begun from Kuşaklı rather than the eastern side of the Kerkenes Dağ we would have walked some 30 kilometres per day without having to cross any difficult terrain. The route was always obvious, although sometimes with alternatives, and we often seemed to be following broad pre-modern tracks. In the later second millennium much would have been forested with cleared roads and tracks. Had we walked the same route in reverse we would have seen the peak of Kerkenes from the first high ridge after leaving the King's Gate at Ḫattusa. The distinctive and imposing landmark would have loomed ever closer from each ridge thereafter. Somewhere, in exceptionally clear conditions even perhaps from that same first ridge, the perennially snow-covered peak of Erciyas Dağ would have appeared more or less directly in line. If the Great King, for war or peace, royal emissaries or merchants were setting out for the Kayseri region, whether to proceed through the Cilician Gates to the Mediterranean ports or through the Anti-Taurus mountains to the Amuq and thence to North Syria and Mesopotamia, the direct line to be followed was highly visible and never in doubt. During the Spring Festival, as noted above, the Great King spent the night at Zippalanda, surely Kuşaklı Höyük. On the following morning, the party went up onto Mount Daha, surely Kerkenes Dağ. After celebrating the mountain, the royal party proceeded in a vehicle to Ankuwa. Ankuwa is almost certainly to be identified with the excavated site of Alişar Höyük by tablets dating to the Middle Bronze Age. From the southern ridge of Kerkenes, below the barren tor, a broad road leads gently down the hillside in the direction of Çadır Höyük some 10 kilometres distant. About as far again in the same direction lies Alişar/Ankuwa. This road was certainly in use in the later Middle Iron Age when an imposing capital was constructed on the Kerkenes Dağ. It leads out of what we have named the Cappadocia Gate, the most imposing of the seven city gates.

The geographical topography of the two big festivals was therefore more or less within the territory of Hattian Hattus(a) (during the period *kārum* Ib Ankuwa was governed by a prince who rebelled

<sup>33</sup> See above, note 8.

<sup>34</sup> This identification was suggested by Gurney 1995, commenting the survey by Summers, Summers, Ahmet 1995. The geographical situation and the recent archaeological operations conducted by the Università di Firenze are in favour of this.

against Ḫattusa)35. It included Arinna and Zippalanda, two of the three holy towns of the Old Kingdom, Nerik (mentioned together with the other two in an Old Hittite manuscript of the *Laws*, § 50)36 being lost until the second half of the 13th century. The redaction of the *nuntarriyašḫaš* Festival of the time of Tutḫaliya IV prescribed a celebration for the Storm-God of Nerik at Katapa on the 14th day (Nakamura 2002: 21, l. 12; 61, l. 3').

The basic rites: the filling and opening of the pithos, the journey of Zitḫariya's Hunting bag (a tutelary god of kingship), perhaps also the connection of this god with the *ḫešta*-house and the cult of the ancestors, were Hattian.

The restyling of the Spring and Autumn festivals at the time of Suppiluliuma and Mursili II extended the celebrations over many days, introducing major Hurrian cults according to the customs of the towns of Halab and Hattarina.

For Tutḫaliya IV, at the end of the Empire period, the basic cultic prescriptions were still those of Ḫattusa, Arinna and Zippalanda, going back to the formation of the Hittite state: «Those cultic prescriptions (*A-WA-AT* DINGIR*LIM*) which the king of Tarḫuntassa provides, they are arranged according to the prescriptions concerning Ḫattusa, Arinna and Zippalanda» (*Bronzetafel* III 61-64).

#### *Bibliography*


<sup>35</sup> KTK 10, ll. 3-8, Larsen 1972. According to Kryszat (1999: 130), the Assyrian document OIP XXVII 1, ll. 1-2, has to be read: 'The seal of Anitta, prince (*ru-ba-e*) of A(n)kuwa'.

<sup>36</sup> Note that the priests (LÚSANGA) of Arinna, Hattuša, Zippalanda and Nerik are listed in this sequence already in a Hattian text, KUB 40.60 + KUB 29.8 (Soysal 2007: 654).

#### 24 *ALFONSO ARCHI*


Larsen M.T. 1972, *A Revolt Against Hattuša*, «Journal of Cuneiform Studies», 24: 100-101.

Maul St. 2013, *Die tägliche Speisung des Assur (ginā'u) und deren politische Bedeutung*, in L. Feliu, J. Llop, A. Millet Albà, J. Sanmartín (eds.), *Time and History in the Ancient Near East. Proceedings of the 56th Rencontre Assyriologique Internationale at Barcelona*. *25-30 July 2010*, Winona Lake: 561-574.

McMahon G. 1991, *The Hittite State Cult of the Tutelary Deities* (Assyriological Studies 25), Chicago. Nakamura M. 2002, *Das hethitische* nuntarriyaš*ḫ*a-*Fest*, Leiden.


# SOME HURRIAN CULT CENTRES NORTH OF THE TAURUS AND THE TRAVELS OF THE QUEEN

# *Massimo Forlanini*

#### *Abstract*

Whereas the main towns of the Cilician plain are never mentioned in old Assyrian sources, their names begin to occur in the Hittite texts in the second half of the 17th century BC. The conquests of Ḫattusili I and Mursili I imply an early control of eastern Cilicia, a region that was later lost during the reign of Ammuna I, becoming the kingdom of Kizzuwatna. The great conqueror Tutḫaliya I (II) made a treaty with Šunaššura, the last independent king of Kizzuwatna, but, later in his reign, he was in control of that country and had a Hurrian (Kizzuwatnean) queen. Since then, we notice a remarkable growing of the influence of Hurrian culture and cults in the official Hittite establishment. From a geographical standpoint the spreading of Hurrian cults affected places of the Upper Land, like its capital Samuḫa and Karaḫna, reaching cities east of Ḫattusa and the new royal residence of Sapinuwa, chosen by Tutḫaliya II (III) in the first half of the 14th century BC. More than one century later, queen Puduḫeba, the powerful wife of Ḫattusili III, probably on her way to Kizzuwatna (or even beyond it, to northern Syria), used to stop over in some Hurrian cult centres on a road connecting the Eastern Taurus passes with the Upper Land, whose names are mentioned in the Hittite votive texts. Starting from this research, the author tries to find in the votive and dreams texts evidence of journeys of the queen towards other destinations (such as Nerik, Tummanna, and Tarḫundašša), witnessed by her presence in various towns lying on the way towards them.

In a recent article (Forlanini 2013), I have attempted to reconstruct the roads net of Kizzuwatna by the mean of Hittite texts of different nature. As a starting point I have recognized that the new localizations by Trémouille (2001) of Kummanni and Lawazandiya in the lower Ceyhan basin, rather than in the Antitaurus area, must be accepted. If the archaeological site of Tatarlı Höyük, excavated by Serdar Girginer, is likely to correspond to Hittite Lawazandiya1 , I suppose that Kummanni, that, according to Shalmaneser III, was located west of Lawazandiya, should be Sirkeli, the archaeological settlement famous because of Muwattalli's relief. In fact, not only its position and importance, but also the special link, witnessed by Hittite evidence, between king Muwattalli and Kummanni2 make this site suitable for the main Kizzuwatnan religious centre.

\*\*\*

A road must have connected central Kizzuwatna through an Antitaurus pass, likely that of Gezbeli (with its Hanyeri relief), with the Upper Land, descending to the area of Fıraktın and then running northwards in the direction of Samuḫa3 . Although Hurrian cult centres were established by the Hittite kings and queens of the New Empire all over Anatolia, it seems that Samuḫa4 was a special step in the process of diffusion of those cults, since the time of Tutḫaliya I, who brought there the cult of the Goddess of the

<sup>1</sup> For Tatarlı Höyük see Girginer 2011. On Lawazandiya (and/or Luḫuzzandiya) see Trémouille 2001, *passim*. Cf. Lebrun 1979, 1981; Wegner 1983; Forlanini 2004, 2013: 7 f . 2 Otten 1981: 15; Bittel 1970: 20 f.; Houwink ten Cate 1983-1984: 68; Ünal 1997: 525; Singer 1996: 193 n. 444;

Singer 1998: 540. With reference to CTH 382 (KBo 11.1) and KBo 22.11.

<sup>3</sup> Forlanini 2013: 20-24.

<sup>4</sup> On the probable localisation of Samuḫa at Kayalıpınar: Müller-Karpe 2000: 363 f. Cuneiform tablets discovered in this main archaeological site have been published by Rieken (2009; 2009a) including a unique historical Hurrian text. On Samuḫa in the Hittite texts: Lebrun 1976. A location of Samuḫa on the Kızılırmak, and not on the Euphrates, was suggested by Goetze and confirmed by a text from Maşat mentioning the neighbouring Karaḫna, identified by Freu with Carana/Sebastopolis (cf. Forlanini, 1979: 181-182 n. 84; Gurney 2003: 125; Wilhelm 2003; de Martino 2009; Barjamovic 2011: 151-154).

#### 28 *MASSIMO FORLANINI*

Night from Kummanni5 . Other Hittite towns having a partially Hurrian pantheon6 my have received the new deities from this irradiation point, I mean Karaḫna (Ḫebat, Šarrumma, Šawušga), Ḫeššašḫaba (Teššub, Šawušga), Kuliwisna (Teššub, Šawušga), Ankuwa (Ḫebat, Šarrumma, Šawušga), Zippalanda (Teššub), Kadaba (Ḫebat, Šawušga), Šulubašši (Ḫebat, Šawušga)7 , and above all Sabinuwa, founded as a royal residence and a Hurrian cult centre by Tutḫaliya II (\*III) and his wife Tadu-Ḫebat8 . A particular role in the area of the Antitaurus was played by the god Šarrumma, specially connected with the mountains9 .

In this article I shall focus on two cult centres of Šarruma north of the Antitaurus, eastern Uda and Urigina, because they are mainly mentioned in texts connected with the activities of a Hittite queen, to be identified with Puduḫeba. It is worthwhile to find an explanation for the importance of these centres only connected with that period: was it their position on the itinerary of the queen's journeys towards her native land of Kizzuwatna?

We have to distinguish between two Udas: a Luwian one, a cult centre of the goddess Šaḫaššara in the Tyanitis, to be located near Emigazi 10, and an eastern one, a sanctuary of Šarrumma and Šawušga, to be probably identified with Fıraktın11. As for Urigina (Šarrumma and Šawušga here too), it should be assumed that this town belonged to the land of Ḫurma, a city located east of Kaniš, probably near Pınarbaşı on the Zamantı Suyu, or further to the east12, and that its name likely survived in Byzantine Aragena and modern Ağırnaz (Taşören), only 8 km. south-east of Kültepe13.

#### *1. Cult centres between Kizzuwatna and the Upper Land*

As mentioned above I have tried to reconstruct a possible itinerary (or parallel itineraries) between Kizzuwatna and the Upper Land through the Antitaurus passes. As main pieces of evidence the following texts have been used.

KUB 18.23: IV 5-7 (CTH 577.1, combined oracle SU, KIN, MUŠEN)14. According to my interpretation this passage shows two different directions from a starting point (Lawazandiya?): on one side Ḫalpa/

<sup>5</sup> See Kronasser 1963; Miller 2004: 259 f.; 2008.

<sup>6</sup> For the cult places of Hurrian deities cf. van Gessel 1998, I s.v. For Ḫebat: Trémouille 1997: 47 f. For Šarrumma: Trémouille 2006; 2009. For Šawušga: Wegner 1981: 157 f. I avoid here to discuss the position of other cult centres of Šawušga in the north, like Tabiniga/Tameninga or Ḫattarina.

<sup>7</sup> For the localizations of these towns suggested by the author see Forlanini 2002; 2008.

<sup>8</sup> The corpus of the Hurrian texts from Ortaköy (Sapinuwa) is presently studied by an international team organized by Aygül Süel, see for instance the papers of the VIU 2011-2012 (Seminar in the Humanities on *Literature and Culture in the Ancient Mediterranean: Greece, Rome and the Near East*), published in Kaskal 10 (2013), 129-183, and the conference *Interculturality in the Hittite Kingdom. Archaeological and Textual Evidence from Ortaköy/Sapinuwa* (Torino 6 June 2013).

<sup>9</sup> Cf. Laroche 1963; Trémouille 2006; 2009.

<sup>10</sup> On Uda: Forlanini 1990; Lebrun 2001; Özcan 2013. Cf. KUB 57. 108 II 12'-17' (cf. Forlanini 1990: 116 f.; Hazenbos 2003: 104-106): for the cult of *Ú-da-a*, including the goddess Saḫḫasarā, also worshiped in neighbouring Tuwanuwa (Uda and Tuwanuwa are mentioned together as the easternmost limit of Arzawa's attack through the Lower Land in KBo 6. 28 obv. 9), are offered animals and flour of the mountain Sarpa, obviously belonging to the town's territory, as pointed out by Lebrun 2001: 331. I have proposed in the past for the Sarpa the Hasan Dağ, too far both from Emirgazi and from classical Hydē (on which cf. Belke, Restle 1984: 174-175) and whose Hittite name could instead have been Ḫarki, since it was called Argaios in Byzantine times, like the Erciyas near Kayseri (Forlanini 2009: 43).

<sup>11</sup> Klengel 1975: 62 (n. 78), pointed out the links between Kummanni and Uda, and between Uda and centres placed in the eastern part of Ḫatti. Therefore, in accordance with Forrer and Laroche (1963: 293-294, 297 map), he preferred a localisation on the road from Ḫattusa to Karkamiš and ruled out the identification with Hyde, assumed by Garstang and Gurney on the basis of KBo 6. 28 obv. 9.

<sup>12</sup> Urigina in the Ulmi-Teššup treaty: CTH 106.B.2 (KBo 4. 10+) rev. 28-32: van den Hout 1995: 48-49. On Pallā lord of Ḫurma: van den Hout 1995: 216-225; van den Hout in fact wondered why this functionary was included among other more important witnesses of the treaty. An easy explanation is that Urigina gelonged to his district. I have located Ḫurma near Pınarbaşı on the Zamantı river and considered her as the new capital of Labarna after the destruction of Kaniš (Forlanini 2007: 264-266, 278-279); other possible locations further east in the plain of Elbistan (cf. Barjamovic 2011: 180-187, with literature). Urigina was then anyway in the Upper Country, and Ḫattusili III, still governor of this region in the time of his brother, erected there a temple for Ištar of Samuḫa: KUB 21. 17 II 7, 27, 36 (CTH 86). See Ünal 1974: 22-24; cf. Laroche 1963: 294. The organization of the cult in Urigina was connected with the expropriation of a 'house' of Arma-Tarḫunda, the former governor of the Upper Land, and in the broken context, also Sippa-zidi, his son, is mentioned: a further confirmation that Urigina belonged to this administrative region, where also Ḫurma was situated.

<sup>13</sup> On Aragena cf. Forlanini 1990: 123 n. 16.

<sup>14</sup> Cf. Forlanini 1990: 113-114; 2013: 21.

Aleppo, on the other side Kummanni, Uda, Urauna, and Azpisna (i.e. Apzisna15). This list seems to go in a south-north direction. I have mentioned the connection between Kummanni and Uda, pointed out by Klengel, and the proposal by Laroche to locate this Uda (the eastern one, having Hurrian cults, not Uda/Hyde near Tuwanuwa) in the Antitaurus area not far from Fıraktın. The sequence Uda Urauna occurs in KUB 40.98, with a north south orientation, in opposite order, Urauna Uda, as it may be expected.

KBo 1.58 1'-5', a list of towns, to be interpreted as an itinerary16: Nenisankuwa, Adunuwa, Apzisna, Sarissa, Samuḫa. Nenišankuwa was situated on a crossroad or in a plain of the Upper Country, not far from Kaniš17. Adunuwa could be equated with neo Assyrian Atuna, and, in this case, perhaps with the Tuna of the Kululu lead strips18. For Apzisna see KUB 18.23 IV 6'. Sarissa (Kuşaklı) and Samuḫa (Kayalıpınar) are not far from each other and connected by an ancient road.

CTH 582, KUB 40.98, an oracular fragment19, according to which somebody performed a ritual in a sequence of towns: Samuḫa, Ḫurma, Urigina, Urauna, Uda, Ellaya, Ḫubisna. It seems that also here the sequence is a chronological one, hence an itinerary. In fact the road links between Ḫurma and Samuḫa are witnessed by Kültepe texts20. Sarissa (Kuşaklı), founded in the 16th century BC, was connected both with Ḫurma and Samuḫa (Kayalıpınar)21. As mentioned above, in the 13th century BC Urigina probably belonged to the land of Ḫurma. We have met above together Urauna and Uda in KUB 18.23. As for Ellaya and Ḫubisna, the sequence here is confirmed by that of Muwattalli's Prayer II 15-23, (Ḫubisna, Tuwanuwa, Illaya, Suwanzana), if we consider that Suwanzana was a town near Tuwanuwa22. As a consequence, if eastern Uda was located near Fıraktın, E/Illaya must be sought between Uda and Ḫubisna and not far from Tuwanuwa, perhaps in a valley of the Taurus range, like that of the Kırkgeçit Dere or, more to the east, near Çamardı.

To summarize the links between the three mentioned lists I give them in a synoptic table:


<sup>15</sup> The correct spelling Apzisna is proved by KBo 1. 58: 3 (see below), by the pantheon-list of Muwattalli's Prayer (II 72-73), and by the occurrence of its Hurrian ethnicon *Apzisnāḫi* referred to a deity (KUB 45. 25: II 8, see Laroche 1976: 51; Haas-Wegner 1988: 394-395). In Muwattalli's Prayer the gods of Abazisna (*A-pa-zi-iš-na*)/ *Apzišna* (hence a better normalization *Abzišna* ?) are the Storm- God (to be probably interpreted as Teššup) and Ḫebat, belonging to the Hurrian cult coming from Kizzuwatna; in the list Apzišna seems to be the last mentioned cult centre of the Upper Land, after Nenišankuwa, Duruwaduruwa and Ikšun(uw)a, sanctuaries of Pirwa north-east of Kaniš, Sullama and Ḫatrā of the country Isuwa, Tegarama, Palliya, Tubaziya (near a lake, north-east of Tuwanuwa in the area of Kaniš, probably not far from Develi (cf. Forlanini 2012: 294), and Kariuna.

<sup>16</sup> See Forlanini 1998: 222; 2013: 22-23. Cf. also Gurney 2003: 125 ('possibly an itinerary').

<sup>17</sup> Cf. Forlanini 2009: 39; 2013: 22 n. 101. Nenisankuwa, Duruwaduruwa and Iksunuwa are mentioned together in the same paragraph of Muwattalli's Prayer as sanctuaries of Pirwa, the deity of Kaneš (CTH 381 A II 62-63, B II 29-30; cf. Singer 1996: 18). Iksun(uw)a is anyway a well known sanctuary of Pirwa and Duruwaduruwa appears in a text from Kültepe as belonging to the direct administration of Kaneš (cf. Forlanini 1992: 179). According to HKM 71 (Alp 1991: 254-259), a letter sent by the GAL LÚMEŠ KUŠ7 ('Commander of the chariots'), who was in charge of the Army of the Upper Land, to Kaššu, the frontier commander of Tabikka, the latter had to join him in Nenisankuwa with the troops of Karaḫna, Isḫubitta and Mount Saktunuwa. The town was situated in the Upper Land, not far from Kaniš and probably in a plain and/or on a cross-road; a suitable location could be in the area of Lake Palas and Sarıoğlan.

<sup>18</sup> Cf. Forlanini 2009: 39; 2013: 23 n. 102. For the Kululu lead strips see Hawkins 2000: Part 2, 503 f., with literature. It can be equated with neo-Assyrian Atuna and Hittite Adunuwa (Forlanini 2009: 39). For other suggestion about Atuna see Bagg 2007: 33-34.

<sup>19</sup> Cf. Forlanini 1990: 114 f.; 2013: 21.

<sup>20</sup> Cf. Kt 92/k 3 (Berjamovic 2011: 152 n. 504, 226-227, *passim*).

<sup>21</sup> On the road system of the Upper Land archaeologically recognized cf. Ökse 2001. On the presence of the "Lord of Ḫurma" in Sarissa cf. KuSa II 13 (Wilhelm 1997, 18-19). On Šarišša: Wilhelm 2009.

<sup>22</sup> According to the inventory KUB 57.108+ II 6-10. (Hazenbos 2003: 104 f.), where the cult Suwanzana is mentioned after that of the goddess Šaḫḫašarâ of 'western' Uda (II 18'-23'). The main sanctuary of this goddess was Tuwanuwa.

#### 30 *MASSIMO FORLANINI*


Why, as said above, Uda and Urigina are frequently mentioned in Hittite votive texts, mainly connected with queen Puduḫeba, the daughter of Kizzuwatna, as she defines herself in the monument of Fıraktın and on her seal, and as it his described in the Egyptian version of the treaty with Ramses II? Here we must connect places where the queen had dreams or made vows and her presence there with specific travels she could have done. In fact we can recognize at least three possible itineraries of the queen.

#### *2. Travels of the queen*<sup>23</sup>*: the Lower land*

We know for example that she visited Tarḫuntassa together with her husband before KBo 4. 10 (obv. 40-47), the Ulmi-Teššup treaty, was written24. Same of the places where the royal couple stopped during that journey, or a similar one, could presumably occur in connection with her dreams or vows, I mean:


<sup>23</sup> Main references in the following chapters for the queen's vows and dreams are: de Roos 2007 (henceforth only: de Roos) and Mouton 2007 (henceforth: Mouton).

<sup>24</sup> «When I, My Majesty went to Tarḫundassa and examined the divine obligations […] But now the king and the queen have made Kurunta king of Tarḫundassa […]». Cf. van den Hout 1995: 34-35; Beckman 1996: 106.

<sup>25</sup> The position of the city Tarḫundassa is still debated: Meydancık Kalesi (Laroche), Dalisandus (Cornelius), the Göksu valley (Hawkins), Karaman (Garstang and Gurney, Gordon), the Kızıl Dağ (Alp), and Hatıp (Bahar) have been proposed (cf. with bibliography). Cf. Alp 1995. It is likely that she was located in the area between Karaman and the Suğla Lake. Cf. Doğan-Alparslan, Alparslan 2011: 94.

<sup>26</sup> The province of Ušša bordered to the south on the Ḫulaya River Land (i.e. Tarḫundassa). The Ḫulaya is generally equated with the Çarşamba river and therefore Ušša must have been located north of it in the direction of Konya, but Ikkuwaniya was not included in her territory. The frontiers are described in the Ulmi-Teššup treaty (van den Hout 1995) and in the Bronze Tablet (Otten 1988); on them: Dinçol *et al.* 2000; Freu 2008: 173 f.

<sup>27</sup> Cf. Trémouille 2006: 203 n. 35 (also on the presence of a king, probably Urḫi-Teššup, at Ušša, according to CTH 703).

<sup>28</sup> CTH 255.2, edition: von Schuler 1957: 8-21 ('Instruktion für Obere'). Cf. Houwink ten Cate 1992: 268; Klengel 1999: 337 n. 64; Lebrun 2001: 329.


If all these places were connected with a journey of the royal pair to Tarḫundassa, we could even try to understand their itinerary: from the Tyanitis (Anasiba), through the district of Ḫurniya and Ikkuwaniya (Saḫḫaniya and Lânda), as far as Ussa on the frontier of Tarḫundassa.

#### *3. Travels of the queen: the north*

A journey of the king to Nerik and beyond it to Tummanna is mentioned in the oracular text KUB 46. 37 Vo. 6: URU*Ne*-*ri-ka4 pa-iz-zi pa-ra-a-ma* URU*Tu*-*ma-an-na pa-iz-zi*<sup>34</sup>*.* The presence of the queen in the far away Tummanna shows that the royal pair paid there an official visit, whose reason could have been a meeting with its king, perhaps on the occasion of his enthronement. A king of Tummanna occurs in the texts mainly because his gifts to sanctuaries of north-western Hittite provinces are registered. We do not know his name, but have suggested that the king Kassu known from a bulla of Nişantaş could have been a king of Tummanna35, and, in this case, he should have been the man appearing in the dream of the queen. Possible stages of this journey or perhaps other similar travels of the royal couple are mentioned in votive texts, but of course they could have been connected only to travels to Nerik:

• Utruna. A town were rituals of Nerik were performed, when Ḫakmis too was lost36. Dream of the Queen in Utruna and vow to Šaušga (KBo 27.60 5' f., de Roos, 146-148, Mouton Nr.105, 280-281); another dream of the queen in Utruna (KUB 48. 121, de Roos 214-215; Mouton Nr. 113, 287-288). Vow mentioning the transfer of the god from Utruna to Nerik (KUB 48. 119 9', de Roos, 208-213).

<sup>29</sup> The religious list of KUB 57.87 obv. I starts with Ušša and the Ištar of Lânda. But also KUB 17.19 I 6 f. could probably link Ussa with Lânda and Lušna (Lystra/Hatunsaray), whereas a connection between Ušša and Lušna may be witnessed by KBo 54. 278: cf. Forlanini 2013: 17, n. 73.

<sup>30</sup> On the evidence about Kun(n)iyawanni, the deity of Lânda, cf. van Gessel 1998: I 262. Since this theonym contains the Luwian ethnic suffix, -*wanni*, the first part must be a place name, i.e. Ikkuwanniya (Iconium, Alaeddin Tepe in Konya), with the lost of the prothetic vowel because of the accent shift. In fact the two towns must have been very close to each other. The other deity of Lânda, an Ištar, remind us of the Ištar figurine found in the Karahöyük

of Konya (Alp 1974). 31 In the Prayer of Muwattalli (CTH 381 II 30-31, Singer 1996: 16, 36) the pantheon (with a Mighty Goddess, dMA<sup>Ḫ</sup>, and a Storm-God) of Saḫḫaniya follows those of Ḫurniya and Zarwiša and precedes those of Paḫtima (probably in the district of Parsuḫanda), Saḫḫuwiya and Mallitaskuriya (on the border of the provices Kissiya and Turmitta, north of the Tuz Gölü). These geographical connections are confirmed by the religious list of KUB 57.87 obv. I, where the gods of Šaḫḫaniya (Tarupšani, Muwatti, Bibira, and MAḪ) precede Nawadiyalla (of Zarwiša) and the gods of Ḫurniya; cf. Forlanini 2013: 16-18.

<sup>32</sup> CTH 19 III 30-32: Parminiyas, Parsuḫandas, Mount......., ..]xdas, Iyammas, Wasuwattas,......,....]xandas, and River Ḫulaya. If the list follows here a geographical order, we should place Iyamma between Paršuḫandaš (Acem Höyük?) and the river Ḫulaya (the Çarşamba). For a possible survival of the toponym cf. Ima(ion) (ethn. Imaēnos) that Ramsay 1906: 365, compared with modern Imen/Emen, a village north of Beyşehir; cf. Forlanini 1988: 152. But the position of Imen does not seem suitable.

<sup>33</sup> Cf. Forlanini 1990: 116-119; Hazenbos 2003: 102-107 (+ KUB 51. 23). The deity of Anaseba is the goddess «Queen of the town» (URU-*LIM*, better than «Queen of the town of Ši[....»). Same town (Anašiba=ma) in KUB 16. 71: 8'.

<sup>34</sup> Before the same text we read *abasila* (Vo 1) surely referred to the king ('the king in person').

<sup>35</sup> Forlanini 2014. I have proposed that the king Kaššu of the bulla 158 (Herbordt 2005: 141 and Pl. 12) could be a king of Tummanna, and the last partially visible sign of his country name could be a NA; anyway no other contemporary kingdom is likely to have had a king with this name, that of the most important and sacred mountain of Tummanna. Moreover, according to Strabo (Geogr. 12.3.25), Gassys was still in his time a typical Paphlagonian name.

<sup>36</sup> Cf. del Monte 1978: 191-192, on the inventory KUB 57. 100 (other edition: Hazenbos 2003: 15 f.).

#### 32 *MASSIMO FORLANINI*


# *4. Travels of the queen: Kizzuwatna and northern Syria*

The examples above could strengthen the idea that the presence of the queen along a route connecting the Upper Land with Kizzuwatna may be connected with journeys to the latter land (perhaps she was bound to Lawazandiya and her family's estates), or even as far as northern Syria, towards Karkemiš, or also in the direction of the Egyptian frontier, when travelled with her daughter for the first part of her way to meet Ramses II39. In fact the evidence shows that she reached Izziya, Class. Issos (Kinet Höyük); pehaps this harbour belonged to the estates of her family, or was only a step towards Northern Syria, since the presence of the queen at Alalaḫ and Mušuniba and Karkemiš is attested as well. Did she reach Izziya (Issos) in order to board on a ship bound to the mouth of the Orontes40, or to cross the Amanus range on a southern pass-road instead of taking the northern one, that of the Adalur pass? As places of her dreams we find Uda and Urigina in the Upper Land and Laiuna probably in Kizzuwatna, but she also frequently made vows to Šarrumma of Uda and Urigina while residing in other towns.


<sup>38</sup> Forlanini, o.c.

<sup>39</sup> Cf. Freu 2008: 235-237.

<sup>40</sup> Perhaps to the harbour on the Oronte's mouth near Sabuniye Höyük (cf. Pamir 2005: 70-72; Oruç 2013: 61 f.).

<sup>41</sup> For this suggested position: Forlanini 2007: 264-266. On Ḫurama in the old Assyrian texts: Barjamovic 2011: 180-187.

<sup>42</sup> Mamma/Ammamma?, cf. Christiansen 2006: 134-135. 43 Bo 3342. Cf. Klengel 1975: 61-62.


<sup>37</sup> Cf. de Roos, o.c. 24-26, who remarked that Manayara was located at a short distance from Kudupsassi. I have suggested (1992: 286-287) that Kudupsassi and Kutupitašša, mentioned in a text from Maşat, are one and the same place and therefore Manayara could be a defective writing for Manaziyara, a town in the province of Ištaḫara, not far from Ḫakmiš; moreover, the town […]*u-pu-wa-iš-ša* (on this reading see already Lebrun 1981: 106 n. 15) mentioned in the same context can only be restored as Pisḫabuwaissa, a town on the road to Nerik (cf. KUB 40. 106: II? 2; Forlanini 2010: 120-121).


### *Bibliography*


<sup>45</sup> Probably between the Cilician Gates and Adana (Forlanini 2013: 18-19). On this important Cilician cult place see Miller 2010: 510-511. 46 The identification of Izziya with Issos, I have proposed because of KUB 56. 15 obv. II 15' (Forlanini 1988: 147;

2001: 553-554), has been favourably accepted (cf. for instance: Gates 2001: 138; Archi 2012a: 45-46; 2012b: 417-419). Bo 4889 (KUB 40. 2) has been transcribed and translated by Götze 1940: 60 f.; Murat 2009: 161-164; Chrzanowska 2011. Cf. also Prechel 1996: 120-122. It is a late Hittite copy of a foundation act concerning the establishment of the Išḫara cult in Kizzuwatna, including an aetiological myth; this act goes back to an old tablet issued by a certain Talzu, probably an earlier king, or even the first Hurrian king of Kizzuwatna, and was confirmed by Šunaššura, the last king of Kizzuwatna before the dynastic union with Ḫatti. The presence of a late copy of a Kizzuwatna document at Ḫattusa can be due to the activity of queen Puduḫeba, the daughter of the high priest of Lawazandiya, who probably inherited the control or the property of those towns, villages and agricultural areas, including Izzi[ya] (obv. 27). Also the case of Šukku and a 'man of Ugarit', concerning a cargo damaged by a collision (RS17.133, cf. Dinçol *et al.* 2000: 9-10), where queen Puduḫeba was involved in the judgement, may have happened in Izziya.

<sup>47</sup> This vow shows that Piyamaradu was living and dangerous for the Hittites in the time of Ḫattusili III.

<sup>48</sup> Kallazu in the texts from Alalaḫ and Kallas(s)u in those of Mari (Groneberg 1980: 130; Belmonte Marín 2001: 151). In the Mari period it was a sanctuary of the Storm-God. Cf. Trémouille 2006: 204, n. 40.

<sup>49</sup> Del Monte, Tischler 1978: 276; Klengel 1995: 132-133, n. 46; Röllig 1997; Belmonte Marín 2001: 201-202. The military itinerary of KBo 8. 35 obv.? (Klengel o.c.) shows that Mušuniba was on the way towards Tunip for a Hittite king (coming from Alalaḫ?).

<sup>50</sup> In the text only Ši[… can be read. The river Šidarpu occurs in lists of Hurrian deities connected with the cult of Teššub and Ḫebat, also those of Aleppo (del Monte, Tischler 1978: 550). A connection with the river Šamura (likely the Šamru of the Šunaššura Treaty, the class. Sarus, today Seyhan), witnessed by KBo 15.45 3-4 and KUB 25.44 II 16', and the mountains Agulliri and Kallištabi would perhaps orientate us towards Kizzuwatna. According to KUB 56.31 IV 13' the queen made this vow in *I-al-an-ta*, a strange spelling for Iyalanda (in Lukka) as remarked by de Roos; if anyway we accept to understand Iyalanda, then we must look for a different river name, perhaps Ši[yanti].

<sup>51</sup> On Tubâ (Umm el-Marra?): Catagnoti 1992; Bonechi 1993: 107-109; Belmonte Marín 2001: 291.

<sup>52</sup> Cf. Klengel 1975: 61-62. Mount Šabaraššana (Akk. Zabarašna) was the Hittite name of the Cilician Gates, cf. Forlanini 2013: 15 f. Does it mean that the queen came to Karkamiš through this mountain pass, or, rather, that she intended to take this road on her way back to Ḫatti and wanted to appease the goddess of the pass?

#### 34 *MASSIMO FORLANINI*

Barjamovic G. 2011, *A Historical Geography of Anatolia in the Old Assyrian Colony Period*, Copenhagen.

Beckman G. 1996, *Hittite Diplomatic Texts* (Writings from the Ancient World 7), Atlanta.


Garstang J., Gurney O.R. 1959, *The Geography of the Hittite Empire*, London.


#### 36 *MASSIMO FORLANINI*


# ADAPTING THE RITE TO TIME AND SPACE: THE HITTITE METEOROLOGICAL CEREMONIES

*Francesco G. Barsacchi*

#### *Abstract*

This paper focuses on the particular Hittite religious expression represented by the celebration of festivals connected to the natural phenomenon of thunder. Interpreted as a manifestation of the Storm-God, thunder was perceived in Hittite religious thought as potentially dangerous for the wellbeing of the king. It demanded therefore the execution of particular rites. The texts describing these ceremonies are collected under the numbers 630 and 631 of the *Catalogue des Textes Hittites*. They show several interesting features, concerning both the content and the internal structure of the compositions, which reflect in my opinion the attempt to adapt a standard ritual practice to an unpredictable event such as thunder.

# \*\*\*

#### *1. Introduction*<sup>1</sup>

Among the hundreds of fragments of festival descriptions unearthed in the Hittite capital Ḫattusa, a particular group of texts is of special interest, casting light on a ritual praxis, i.e. the celebration of festivals and rites directly related to the meteorological phenomenon of thunder, which has no direct parallel in other Near Eastern cultures, and provides interesting insights into Hittite religious thought. The fragments pertaining to this typology of texts are currently listed under the numbers 630 (*La lune et le tonnerre*) and 631 (*Fête du tonnerre*) of the *Catalogue des textes Hittites*. This study aims at analyzing a few elements relating to the content and the internal structure of these tablets, in an attempt to underline some interesting features concerning their redaction and organization. They represent, in my opinion, a particular category of *Sammeltafeln*<sup>2</sup> , being divided in thematically related sections where several festivals connected to thunder or different versions of the same rite seem to be described.

In common with every natural phenomenon of particular violence that lies beyond the possibility of human control, thunder was perceived in all ancient Near Eastern religions as a *numinous*<sup>3</sup> power. As a manifestation of the main male god of the pantheon, the Storm-God, it causes in Hittite thought a particular religious fear which demands to be controlled and exorcised by means of appropriate ritual practices. These rites are aimed at preserving the ritual purity of the king, endangered by what is interpreted as a clear sign of divine anger, while establishing at the same time a form of religious con-

literature. 3 For the concept of *numinous*, see Otto 1936.

<sup>1</sup> This article presents some of the results of my doctoral dissertation *Le cerimonie ittite del tuono: edizione critica di CTH 630 e 631*, which was submitted to the University of Florence in April 2014 and is currently being revised for publication. The research was conducted at the Universities of Florence and Würzburg, under the joint supervision of Prof. Franca Pecchioli Daddi, Prof. Giulia Torri and Prof. Daniel Schwemer. To all of them goes my deepest gratitude. It is a particular pleasure to contribute with this short article to a volume dedicated to Prof. Franca Pecchioli Daddi, whose inspiring lessons I still recall with great delight. I thank Giulia Torri, Michele Cammarosano and Carlo Corti for reading and discussing previous drafts of this work and for their valuable suggestions. I am, of course, responsible for any and all errors or mistakes. 2 For a typological classification of Hittite *Sammeltafeln* see most recently Hutter 2011: 115-120, with further

trol over the phenomenon4 . From the Old Hittite period, they are written down in double-columned tablets, defined in the colophons (preserved in KBo 17.74+, rev. IV 43', KBo 17.75, rev. IV 57'' and KBo 20.61+, rev. IV 14') simply as 'tablets of thunder', DUB/*TUPPI tetḫešnaš*, characterized by a strong uniformity in content and structure.

Besides many fragments currently classified under CTH 631 which cannot be placed with certainty due to their fragmentary state, three compositions clearly stand out as particularly significant:


To these it is possible to add VSNF 12.10, certainly belonging to the same typology but very badly preserved.

CTH 630, on the other hand, collects the fragments that can be connected to a particular thunder festival, celebrated when an outbreak of thunder interrupts the celebration of an ongoing ceremony, defined both in the incipit of the composition and in the preserved colophons as «festival of the Moongod», EZEN4 DEN.ZU5 .

KBo 17.74+, published by E. Neu in 1970 as *Ein althetitisches Gewitterritual*<sup>6</sup> , is the only text of this group which has been properly edited. All the compositions belong to the textual typology of the socalled «festival descriptions», *Festbeschreibungen*, but present some peculiar features. They all describe ceremonies closely related to an unpredictable event such as the sudden thundering of the Storm-God. As evident from the opening description of these texts, the development and the articulation of the ceremonies depended on the position and the action performed by the king at the very moment of the acoustic perception of the thunder.

While a standard ritual procedure is recognizable, the particular rites that mark the various moment during the ceremony were therefore subject to modifications according to time and space.

The formula which introduces the texts begins as a rule with a temporal clause which focuses on the action carried out by the king or the royal couple at the moment of the thundering and takes the form: *mān* LUGAL*-uš…*(EGIR*-an*) DIM/U *tetḫa*(*i*), «when the king […] and the Storm-God thunders […]». The sentence is also attested in the reverted form: *mān* DU *tetḫa*(*i*) (EGIR*-an*) LUGAL-*uš*, «when the storm-god thunders, the king […]».

#### *2. The Old Hittite tablets*

### *KBo 17.74+ // KBo 17.11+*

In the Old Hittite text KBo 17.74+ and its duplicate KBo 17.11+7 , this opening formula occurs three times.

As already recognized by G. Wilhelm8 , the document constitutes a *Sammeltafel* where two distinct, albeit similar in content, meteorological ceremonies are collected: «wir haben […] mit einem Akt kompilierenden Sammlung zusammengehöriger Rituale zu tun» (Wilhelm 1995: 387). The first one stretches from line 1 to line 25 of the first column of the obverse, and the second one, separated from the other by a double paragraph divider, from line obverse 28 onwards. Between the two descriptions, a short redactional insertion of only three lines summarily describes the procedures the king has to follow in the case of a second outbreak of thunder when the first rite is already finished. In this respect,

<sup>4</sup> The clearest example of the religious significance attributed by Hittites to thunder is found in the composition generally known as Muršili's *aphasia*, CTH 486, where the illness of the king, interpreted as the external manifestation of a condition of impurity, is attributed to the sudden thundering of the Storm-God. Cf. Neu 1970: 65; Houwink ten Cate 1992: 92. 5 *Infra.*

<sup>6</sup> Neu 1970.

<sup>7</sup> Transcribed by Neu 1980. With respect to KBo 17.74+, KBo 17.11+ appears to be a slighter older redaction, as shown by the general aspect of the *ductus*, e.g. the density of the script and the spatial organization of the text, as well as by the presence of the older version of the signs DA and IT, a feature, however, which cannot in itself determine with certainty the relative dating of two texts. Both redactions must then be considered OS texts, and the differences between them are not such as to rule out the possibility that they may depend on different scribal habits. Two different scribes could have been responsible for the redaction of the two copies. 8 Wilhelm 1995: 383, 387.

the short succession of ritual acts described cannot be considered an independent rite but rather an appendix of the first ceremony, as suggested also by the fact that only a single paragraph line divides the two sections of the tablet9 . The passage reads as follows:

*\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_*

KBo 17.74+ // KBo 17.11+, obv. I

25 *ma-a-*[*an* (*ḫa-at-kán-zi*)] DIM*-ša* ⌈ *ti*⌉ *-it-ḫa* LUGAL*-uš lu-ut-ti-ia-aš pé-ra-an* 26 *a-r*[*u-wa-iz-z*]*i* NINDA*ḫar-ši-in pár-ši-*⌈ *ia*⌉  *ta lu-u*[*t-t*]*i-ia-aš da-a-i* DUG*iš-pa-an-tu-zi* 27 *da-a-*[*i t*]*a ši-pa-an-ti* LUGAL*-uš nam-ma a-*[*r*]*u-wa-a-iz-zi*

«When [they close] and the Storm-God thunders, the kin[g bow]s before the [win]dow. He breaks a thick bread and places (it) by the window, he takes an *ispantuzzi-* vase and makes a libation. The kings bows again».

\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_

Three main actions are prescribed in this section: a double bowing gesture before a window, before and after the ritual sequence, a bread breaking and a libation. These three basic acts form in a certain sense the core rite in all ceremonies connected to thunder, fully expressing their meaning and purpose: the need to appease the angry god with offerings and libations and to protect the king from the potential danger perceived in the thunder.

The first ceremony described in the tablet occurs in the building complex known as *ḫalentiu-*. The incipit reads as follows:

KBo 17.11+, obv. I

1 [*ma-a-an* DIM*-aš ti-i*]*t-ḫa* DUMU ⌈ É⌉ .[G]AL *píd-da-a-i* (*…*)

«[When the Storm-God thun]ders, a Palace attendant runs […]».

The hierophany represented by the thunder leads to the interruption of every ongoing action, and to the beginning of a particular sequence of rites which constitutes the thunder festival. The first ritual act performed by the king is a bowing gesture before the window (*peran aruwaizzi*), followed by a libation. It is an important preliminary act which introduces the celebration of the proper ceremony. The window, hittite GIŠ*luttiya-*, seems to have a particular religious meaning in all rites connected with thunder. It is one of the sacred places of the house which receives offerings during many Hittite festivals and rites, defining the symbolic limit between inside and outside10. In thunder ceremonies, it represents an opening towards the external space where the sacred has manifested itself in the thunder, assuring communication between the king and the enraged Storm-God. As observed above, the same action, a bowing towards the window, is prescribed in the short insertion between the two sections separated by the double paragraph divider. The following acts described in the first text of the tablet consist in standard drinking ceremonies, libations and the breaking of bread. Among the deities which receive offerings, the couple constituted by the Storm-God and the Hattian god Wašezzili, most likely to be identified with the Storm-God of Zippalanda11, undoubtedly occupies a prominent position, which is not surprising in view of the nature of the text. The same can be observed in the other compositions of the *corpus*.

In line obv. I 24, the text explicitly states that the ceremony in the *ḫalentiu-*building is ended.

The second text contained on the tablet, as correctly pointed out by G. Wilhelm (Wilhelm 1995: 387), shows a much larger degree of ritual elaboration. It describes a festival that lasts for three days and takes the form of daily (cf. Obv. II 61: *šiwatti tuḫḫušta*, «in the day (it) is finished») and nightly (cf. rev. IV 26': *išpanti tuḫḫušta*, «in the night (it) is finished») rites. The ceremony begins when the king is outside, on the chariot *ḫuluganni-*.

<sup>9</sup> Thus also Wilhelm 1995: 386, according to whom «händelt es sich bei dem 2. Ritual eher um eine Verhaltenvorschrift für den König als um ein wirkliches Ritual».

<sup>10</sup> For the symbolic value of the window in Hittite literature and its use as a literary *topos* see, most recently,

Christiansen 2007: 143-152. Cf. also Haas 1994: 276ff.; Popko 1978: 44-48. 11 See Pecchioli Daddi 1998: 25. The observation of Yoshida 1996: 324, according to whom the couple: «(Wettergott) Wettergott von Zippalanda hat nach bisherigem Wissen sonst nichts oder kaum etwas mit Gewitter [*tetḫeššar*] zu tun und erhält kein Opfer in Gewitterfesten», must thus be largely reconsidered.

#### KBo 17.74+, obv. I

28 *ma-a-a*[*n* DIM*-a*]*š ti-it-ḫa* LUGAL*-uš* GIŠ*ḫu-lu-*[*ga-an-n*]*i-az kat-ta ú-iz-zi*

*«*When [the Storm-God thunders, the king] goes down [from the] *ḫulu*[*ganni-*] chariot*»*.

Immediately after hearing the thunder, he gets down from the chariot, he bows to the ground and makes a libation. Then he reaches a city whose name is not mentioned in the text, thus leaving open the possibility that the rite was conceived in order to be adapted to different situations, even though the name of the city might as well be understood. The continuation of the rite, which starts the day after the thundering and includes sacrifices, libations and drinking ceremonies performed by the royal couple, will not be analyzed in detail here. I will merely underline the importance of the ritual practice, described in the lines obv. I 39-42, centered on the manipulation of the object called *ḫulliš*, probably to be identified with juniper wood12, which is thrown into water and then used to purify the king. Closely connected to this operation is the ritual use of the substance called *tuḫḫueššar*, which is given to the king, who then seems to cut it off (*tuḫš-*) 13. The meaning of the whole operation is clearly a ritual purification of the king, whose state of holiness is being somehow endangered by the god's wrath. The same operation is described in VSNF 12.10, rev. IV 16'-21'14.

#### *3. The Middle Hittite tablets*

#### *KBo 17.75*

Compared to the Old Hittite *Sammeltafel* KBo 17.74+, KBo 17.75 presents some more problematic features. It is a later text (MH/MS), as shown by its Middle Hittite script15 as well as by its grammatical and linguistic features16. Like the second festival description collected in KBo 17.74+, the rite begins «when the king goes out and the Storm-God thunders»:

KBo 17.75, obv. I


5 *na-aš da-ga-a-an* [*a-ru-wa*]*-a-iz-zi*

*«*When the king goes out and the Storm-God thund[ers] the king gets down from the *ḫuluganni*chariot. The chief of the Palace attendants takes him by the hand and brings him to the garden. A member of the Guard takes the throne and puts it on the ground».

Also in this case, the king is on the *ḫuluganni-* chariot. A GAL DUMUMEŠ É.GAL, a «chief of the palace attendants», takes him to a place indicated by the logogram KIRI6 , a «garden», where the king bows to the ground. No more specific indications as regards the place where the rite takes place are given. This ambiguity, together with the standardized character of many of the rites described, appears as a typical feature of all the compositions collected in CTH 631, which are not to be seen as 'real' descriptions of festivals, but rather as particular scribal works composed with a practical purpose: collecting a general report of the rites to be performed when violent thunder storms, in specific circumstances, endangered the wellbeing of the king.

<sup>12</sup> As already observed by Christiansen 2006: 88. However, cf. also Rieken 1999: 400, who hypothetically trans-

lates the term as Holzkohle, 'coal'. 13 On this substance, see Zeilfelder 2000: 497-508, who identifies it as a sort of aromatic resin of semi-liquid consistency, used in several ritual contexts for purifications. The semantic connection between *tuḫḫueššar* and the substantive *tuḫḫu(w)ai-*, «smoke», already proposed by Güterbock 1964: 106ff. would thus lie in the particular smell emitted by the *tuḫḫueššar.* substance. For a different interpretation, see HW2

: 317; Kloekhorst 2008: 892-893. 14 See now Barsacchi 2014: 245-246. 15 Among other paleographic features, it is interesting to note the presence of the signs AZ with the little ZA

subscribed and E with the vertical wedge in level with the higher horizontal. 16 The presence of the conjunction *nu* instead of *ta*, the predilection for logographic instead of phonetic writings and, from a morphological point of view, the use of the *-ri* form in the present indicative endings of the medio-passive conjugation are the most significant elements that indicate a later phase of redaction. Among the linguistic features catalogued by Melchert 2008: 525-531, as typical of a Middle Hittite phase, we find the use of *andan* with the locative meaning 'inside' and that of the enclitic personal pronoun *-e* instead of *-at* in the 3rd person plural inanimate.

The fragmentary state of lines 10-21 does not allow us to understand how the rite continues. Starting from line obv. I, 22, a new rite seems to be described:

\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_


«When […] and the king g[oes? ] to the bath-house [and] the Storm-God thunders, the foreign guests […] and they lead them out».

The introductory formula closely resembles the formula already seen as typical of thunder ceremonies. In this case, the thunder interrupts a movement of the king to the so-called «bath-house», É.DU10. ÚS.SA, probably together with his court. The lack of the double paragraph line which one would expect as divider between the two parts of the text is striking. The second one could thus be regarded as a description of the ritual actions to be performed when a second instance of thunder interrupts the original rite, making a modification of the normal procedure imperative. The development of thunder ceremonies necessarily shows a high degree of flexibility, dependent as it is on the variation of the meteorological situation. More important than the description of the continuation of the festival, in which the sequence of actions and events was probably well codified from older periods, was the exact account of the modifications in its basic structure necessitated by unpredictable events. The same urgency explains the redactional insert represented by KBo 17.74+ // KBo 17.11+, obv. I 25-27.

The colophon, fully preserved, does not help us in understanding the character of the text, defining the tablet only as «tablet of the thunder», as in KBo 17.74+. It is interesting to note the presence, in line rev. IV 50'', of the expression *ANA* GIŠḪUR*-kán ḫandan*, which seems to suggest that the copy has been redacted on the basis of (or in accordance with, according to the meaning attributed to this controversial formula17) a wooden record. This is not without significance, in that it could indicate that the ceremony described was actually performed at the time of the redaction of the tablet18.

#### *KBo 20 61+*

KBo 20.61+, another Middle Hittite composition, displays perhaps the most peculiar features among the texts collected in CTH 631. To the fragments of the tablet I now add KBo 17.77, which is attributed to the shelf list KUB 30.57+KUB 30.59 (CTH 276)19 in the online *Konkordanz*20, but which represents, in my opinion, the colophon of this text, as seems to be confirmed by both the handwriting and the outline of the fragment21. It is in indirect *join* with the rest of the fourth column, which is lost in lacuna after a few lines. The peculiar articulation of the colophon22, where different incipits are mentioned, confirms the fact that the composition is in some sense a compilation.

The ceremony begins «when the king and the queen sit in the great assembly and the Storm-God thunders»:

KBo 20.61+, obv. I


«When the king and the queen sit in the great assembly and the Storm-God thunders, they make the sons of the king and the daughters of the king, the dignitaries, the foreign guests and those who sit in front of the king rise, and the "staff-men" lead them out».

<sup>17</sup> For different interpretations of this controversial expression see, among others: Singer 1983: 41; Starke 1990: 459; Marazzi 1994: 144-148, 153; Schwemer *forthcoming*; Waal 2011: 22-26, all with further literature. 18 As correctly pointed out by Lorenz 2014: 459-474.

<sup>19</sup> Where the composition seems to be actually mentioned, in ll. 8'-10' of the left column. Cf. Dardano 2006: 49. 20 Košak, hethiter.net/: hetkonk (v. 1.91).

<sup>21</sup> I am indebted to Michele Cammarosano for the collation of the fragments carried out for me in September 2013 at the Museum of Anatolian Civilizations of Ankara. 22 *Infra.*

#### 42 *FRANCESCO G. BARSACCHI*

The 'great assembly', whose celebration is documented from the Old Hittite period, constitutes a particularly significant moment from a religious and a symbolic point of view23. The interruption of the ceremony caused by the thunder needs a ritual response, which takes the form of a thunder festival. Typically, the first sequence of rites takes place before the window, where libations and the breaking of bread, accompanied by songs and music, are performed. The fragmentary state of the first column of the text does not allow a full comprehension of the rite, which ends with the distribution of food and beverages to the participants by some categories of palace functionaries, immediately before the standard action of shouting: «curtain!» and the closure of the palace for the night, at the lines obv. II, 16-17.

At this point we find the sentence: «The following day [they celebrate (?)] the festival of the thunder», followed by a double paragraph line, upon which it is possible to read, even though in broken context: […]*ta ḫukanzi*, «they slaughter».

KBo 20.61+, obv. II

16 (…) KUŠNÍG.BÀR


«[There is a shou]t: "curtain"! and they close for the night. […] The day after [they celebrate (?)] the festival of the thunder…] they slaughter».

*\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_*

This passage, where the only attestation of the expression EZEN4 *tetḫešnaš* in a text included in CTH 631 is documented, poses several interpretative problems.

The double paragraph line seems to end the first composition and to signal the beginning of the second one, as usual for Hittite *Sammeltafeln*. The following paragraph, in fact, begins with the standard introductory formula of this category of documents, describing how the Storm-God thunders when the palace is being open, and the king comes from the «inner chamber»:


«When they open the *ḫalentiu-* building, and they draw the curtain, and the king [come]s [from the] inner chamber, but it thunders, [the king] go[es] before the win[dow and] he bows».

*\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_*

The formula is repeated in the colophon, KBo 17.77, where the single compositions collected on the tablet are mentioned according to their respective *incipit*.

The last sentence of the former paragraph, however, clearly seems to indicate a continuation of the rite on the following day. The opening of the palace, and the act of drawing the curtain which opens the second composition in lines obv. II 20-21, constitute a standard morning operation in festivals that last for more than one day24. Moreover, there is a strong continuity between the two parts of the text as regards the rites and the cult functionaries mentioned. Despite the indication on the paragraph line, an animal slaughtering, as far as it is possible to understand, is not mentioned in the following sections of the text.

The same situation can be observed at the beginning of the third column, at lines rev. III 2-7, where the interruption of the rites for the night is followed by what seems to be the beginning of a new, occasional, ceremony, originated by the emergence of the thunder «when they open (i.e. the palace), they draw the curtain and the king goes out». In this case, a single paragraph line divides the two different sections of the tablet:

<sup>23</sup> See Archi 1979: 197-214.

<sup>24</sup> See Lorenz forthcoming*.* 

\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_

KBo 20.61+, rev. III



«T[here is a shou]t: ["curtain"!] and [they close] for the night. The day after they [celebrate (?) the festival of thun[der]. When they open and [they draw the curtain], the king go[es] out […] the Storm-God thunders […]. Then the king gets down [from the *ḫuluganni-* chariot (?) and he bows [to the ground (?)], the ma[n…]».

How to explain then the discrepancy between the use of the double paragraph line and the presence of the second introductory clause on the one hand, which seem to indicate the beginning of a new ceremony, and, on the other hand, the clear hints at a continuation of the thunder festival on the following day?

In my opinion, KBo 20.61+ could be analyzed as the description of a thunder festival extended with the inclusion of two redactional additions. These describe the rites that have to be performed *in the event that* a new episode of thunder, at the beginning of the following day, necessitates a modification in the regular continuance of the ceremony. The scribe, in other words, seems to have decided not to insert in the tablet the description of the entire festival, summarizing its continuation in the phrase: «the following day they [celebrate (?)] the festival of the thunder». The first rites performed in connection with the atmospheric hierophany were probably the most significant from a religious point of view, being aimed at preserving the ritual holiness of the king, and needed therefore to be accurately described. The ritual frame developed around these basic apotropaic rites might well have had a more standard character. If this hypothesis is correct, the following sections of the tablet could be interpreted as the description of the modifications to the original rite necessitated by the sudden emergence of more outbreaks of thunder during the following days of the ceremony. According to the actions performed by the king and his position at the moment of the acoustic perception of the phenomenon, the festival continues in partially different ways.

The *incipits* of the different sections are then repeated in the colophon, represented by KBo 17.77. The uncommon use of the conjunction *našma* between them also seems to indicate that we are dealing with alternative versions of the same ceremony, rather than with entirely different compositions.

14' [DUB xKAM] *ti-it-ḫ*[*é-eš-n*]*a-*⌈ *aš ma*⌉ *-a-an* [LUGAL MUNUS.LUGAL]

15' [*šal-la*]*-*⌈ *i* ⌉  *a-še-eš-ni a-ša-an-zi* E[GIR*-an-na*]

16' [ DIM*-aš*] *ti-it-ḫa-i na-aš-ma-kán* [*ma-a-an* LUGAL*-uš*]

17' [x x (x) ]x *ku-wa-pí tu-un-na-ak-*[*ki-iš-na-za*]

18' [*ú-iz-z*]*i* [ DIM]*-aš-ma ti-it-ḫa-*[*i*]


21' (traces)

«[… tablet] of the thunder: when [the king and the queen] sit in the great assembly [and the Storm-God] thunders, or [when the king…] when he [comes from] the inner chamber but the [Storm-God] thunders, or [when the ki]ng [goes] out [and the Storm-God thun]ders […]».

I am aware that this is only one of the possible explanations for the peculiarities shown by KBo 20.61+, and that several problems, such as the incongruent use of the double paragraph line in the text, still have to be resolved. On the other hand, analyzing the tablet simply as a standard *Sammeltafel*, collecting together three distinct thunder ceremonies would, in my opinion, be misleading, considering the clearly structured daily articulation of the ceremony it describes. One and the same festival description, then, but constantly modified in accordance with the atmospheric situation.

<sup>8</sup> *n*[*a-a*]*š-ta* LUGAL*-u*[*š* GIŠ*ḫu-lu-ga-an-ni-ia-az* (?)]

#### *4. The Moon and the Thunder*

The clearest example of the adaptation of a rite to the contingent situation is represented by the so-called «festival of the Moon and the Thunder», whose fragments are collected under the catalogue number CTH 630. This very peculiar composition is preserved on several tablets, which date to different periods. In particular, it is possible to identify a Middle Hittite version of the festival, represented by the tablet KUB 32.135+, with its duplicate KBo 20.63, and by KBo 23.64+KBo 24.88, and a Late Hittite version, preserved in KBo 25.178 and KBo 21.88+KBo 20.70.

The observation of the textual material allows us to identify more than one redaction of the main festival description. The preserved colophons, all of them mentioning the title of the composition, hint at the existence of at least a version enumerated by tablets and a version enumerated by days, similarly to the observation of I. Singer (Singer 1983: 114) with regard to the KI.LAM festival. The ceremony seems to last for at least three days.

According to E. Laroche (Laroche 1973: 89), the text is the result of the combination of two originally distinct Hattian festivals, realized by the Hittite scribal school at the time of the alleged 'religious reform' of Tutḫaliya IV, whose nature and extent, however, has now been largely revisited25. This would explain, according to Laroche, the false archaisms and the formal and structural contaminations which he finds in the composition, and especially in the long list of Hattian deities included in two of the tablets, KUB 32.135+ and KBo 21.88+. The comprehension of the text cannot be separated, in my opinion, from the broader picture represented by Hittite thunder ceremonies. As we have seen, the ritual development and the organization of these festivals are closely dependent on variable causes.

Strictly speaking, it is correct to say that there is not a 'thunder festival', in the sense of a fully structured ceremony repeatable through time in the same form. Every ceremony of this kind constitutes a ritual response to a specific situation. Far from being just the fusion of two distinct ceremonies, the fragments of CTH 630 constitute the original product of the superimposition of the rites connected with the festival of the Moon-god, whatever this expression indicates26, with the particular apotropaic rites that were typical of thunder ceremonies, whose performance, however, largely depended on non-predictable factors.

Even if preserved in later copies, the festival described in CTH 630 certainly contains elements of ancient Anatolian tradition, as correctly pointed out by Laroche. The celebration of meteorological ceremonies is attested from the Old Hittite period, as testified by KBo 17.74+ and its duplicate. The existence of a Middle Hittite version of the Moon and Thunder festival indicates a textual tradition which dates to the same period and is probably to be connected with an earlier stage of local beliefs. This seems to be confirmed by the two long lists of deities already mentioned, which describe what undoubtedly constitutes a Hattian pantheon, and can be compared with the deities of the 'great assembly' of the KI.LAM festival27.

Unfortunately, the fragments belonging with certainty to this group allow only a partial reconstruction of the festival outline.

From the incipit, preserved in both duplicates KUB 32.135+ and KBo 20.63, we know that the composition describes the rites performed when an outbreak of thunder interrupts the celebration of the festival of the Moon-god, EZEN4 DEN.ZU28.

KUB 32.135, obv. I


<sup>25</sup> See now Cammarosano 2012a: 345ff; 2012b: 14-33. 26 In view of the lunar character of the Hittite calendar, on which see Del Monte 1988, 51-56, it is likely that the 'festival of the Moon' was somehow connected with the festivals of the month, or 'monthly festivals', i.e. the ceremonies performed monthly at every new or full moon, and often mentioned in cult inventories as part of local liturgical calendars. The celebration of a festival of the Month is documented also in the domain of the so-called state cult. Cf. the fragments collected in CTH 591, published by Klinger 1996. 27 Laroche 1973: 83-89. Cf. also Singer 1983: 101-103.

<sup>28</sup> Cf. the writing D30 in obv. I 8.


«When "the washing" for the festival of the Moon-god (takes place)29, the king washes himself, but when the king comes from the bath-house and it thunders, then, if the man of the Storm-god has not yet been engaged, they leave the festival of the [Moon-god] and celebrate the festival of the thunder».

 In particular, the text specifies that the action interrupted by the thunder is a movement of the king from the «bath-house», where he has performed the ritual ablution, expressed by the verb *warp-*, prescribed for the celebration of the Moon festival. It must be a particularly meaningful moment in Hittite ritual praxis, considering that a clap of thunder during an analogous occasion, «when the king goes to the bath-house», originates the celebration of the second ceremony described in the text KBo 17.75.

In this case, the text prescribes that, if the cult functionary known as «man of the Storm-God»30 has not yet been engaged in the celebration, the Moon festival can be interrupted and the thunder festival is celebrated in its place.

If, instead, the «man of the Storm-God» is already present, the celebration of the moon festival and of the thunder festival are mixed together, and the two ceremonies are performed jointly: \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_


«Otherwise, if the Man of the Storm-God has already been engaged, the festival of the Moon-God (and) the festival of the thunder are mixed and they are performed jointly».

*\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_*

In the composition, only this first situation is described. Once again, the standard celebration of the thunder festival is not contained on the tablets, which describe only the adapted version of the original rite.

On the first day, after the canonical libations before the window and the bread offerings for the Storm-God, a bull and a ram are slaughtered and offered. Then the «great assembly» takes place. The indication, in KUB 32.135+, obv. I 30-32, that the king and the queen eat the 'pure sheep' that is usually cooked for the festival of the moon is a clear sign of the complete adaptation of the ongoing ceremony to the new situation modelled by the atmospheric hierophany.

The aforementioned lists of deities which receive offerings during the great assembly are preserved in fragments KUB 32.135+ and KBo 21.88+KBo 20.70, but it is questionable whether the last of these belongs to the same composition. The two lists show in fact several discrepancies as regards the names and the position of the deities mentioned, which, as stated before, certainly belong to the Hattian nucleus of Hittite religious belief. Although some later contaminations can be recognized31, both lists are certainly based on Old Hittite models.

Another element which seems to indicate a substantial continuity in the textual tradition from the Old Hittite period is contained in KBo 25.178, a text belonging to the second day of the festival, as confirmed by the colophon in lines rev. IV 20'-22'. Towards the end of the fourth column, in a fragmented context, ritual actions are performed by the DUMU.LUGAL, 'the prince' in the building defined as the «house of the ZABAR.DAB functionary», literally the «holder of the bronze bowl», probably a building used to collect the ceremonial wine, to be identified with the É.GEŠTIN32. In particular, a libation for a group of seven deities is mentioned33. This operation recalls the one performed by the prince in the Middle Hittite text KBo 20.44+, concerning the rituals performed during the AN.TAH.ŠUM festival in the temple of the Sun-goddess, on the seventh or eighth day of the celebration34. In this case, a partially different group of seven deities receive offerings from the prince in the É.ZABAR.DAB building, in lines rev. III 7'-10'. Both texts are testimonies of the same rite, and they show strong similarities with what emerges

<sup>34</sup> Published by Yoshida 1992: 121-156.

<sup>29</sup> *Pace* Ünal 1998: 76, the verbal form *warpuwanzi* can't be translated as a present tense 3rd person plural of *warp-*, for which a form *warpanzi* would instead be expected. The verb can in my opinion only be interpreted as a substantivated infinitive in a nominal sentence. 30 On which see Ünal 1998: 67-82.

<sup>31</sup> Laroche 1973: 86-87.

<sup>32</sup> The role of the DUMU.LUGAL in Hittite festivals has been thoroughly studied by Torri 2004: 461-469.

<sup>33</sup> On which see Yoshida 1996: 88-94, in particular 91-92.

from an Old Hittite document, KBo 25.51+35 , currently filed among the fragments of thunder festivals under CTH 631, but probably to be assigned elsewhere. The rites performed by the prince in the «house of the wine», in other words, seem to reflect a ritual praxis which can be related to the oldest Hittite period.

#### *5. Conclusion*

To sum up, the apotropaic rites performed by the Hittites on the occasion of particularly violent thunder storms constitute a very particular form of religious tradition, which has no parallel in other ancient Near Eastern cultures, where thunder as a natural phenomenon has a strong significance in oracular praxis36 but does not seem to be reflected directly in religious worship.

The need to adapt a standard ritual procedure to an unpredictable event such as thunder lies at the basis of the production of a special type of documents. The 'tablets of the thunder', as the compositions collected in CTH 631 are defined in the colophons, constitute in my opinion a peculiar kind of redactional work where, in distinct sections, both the description of the festival connected to thunder and that of the modifications to the original ceremony necessitated by the ongoing meteorological situation are collected. The Old Hittite *Sammeltafel* KBo 17.74+ // KBo 17.11+, with its short redactional insert dividing the two festival descriptions collected on the tablet, could represent an earlier stage of this practice, fully developed in the later, Middle Hittite compositions, entirely dedicated to the description of a single ceremony and its (eventual) continuation or modification. This work reflects the need to have a well-established ritual procedure by means of which to respond to a situation perceived as potentially dangerous, the sudden thundering of the Storm-God, an event with a strong religious significance that cannot be predicted with certainty and is therefore particularly feared.

These rites constitute the reflex of ancient religious concepts probably dating back to the Hattian cultural *milieu*. Indeed, the presence of OH/OS copies indicates an old Anatolian background for these compositions, as confirmed by the relevant position that the dyad: Storm-God-Wašezzili occupies in all the divine lists. The presence of Late Hittite copies of the festival of the moon and the thunder proves the continuation of this festival tradition until the latest phase of the Hittite empire.

Before concluding, a short consideration is in order. All the texts collected in CTH 630 and 631 reflect only the 'official' aspect of the particular form of Hittite religious expression represented by the thunder festivals. However, these ceremonies were also celebrated at a local level. Several cult inventories, in fact, mention the celebration of 'thunder festivals' as a regular activity which took place annually in several cult centers and was, in any extent, part of local religious calendars. In particular, ceremonies connected to thunder are attested in Nerik (cf. KUB 53.21, obv. 10'37), Ḫurma (cf. KUB 56.56, rev. IV 938) and Karaḫna (cf. KUB 38.12, obv. II 9, 2339). Apart from a few scarce indications concerning the goods to be prepared for the festival and the deities involved, nothing of this local religious tradition is known. The absence of the royal couple as main executor of the rites certainly implied some differences in the development of the whole ceremony, which was not centred on the ritual purification of the king but had most likely the character of a seasonal festival40, performed on a regular basis as part of the cult due to local weather deities, and thus not necessarily dependent on contingent circumstances such as the festivals collected in CTH 630 and 631.

#### *Bibliography*

Archi A. 1979, *Das Kultmahl bei den Hethitern*, in *VIII. Türk Tarih Kongresi, Ankara, 11-15 Ekim 1976. Kongreye Sunulan Bildiriler [Communications Presented to the Congress]*, Ankara: 197-214.

Barsacchi F.G. 2014, *Le cerimonie ittite del tuono. Edizione critica di CTH 630 e 631*, Unpublished dissertation, Università degli studi di Firenze.

<sup>36</sup> See the meteorological *omina* of the series *Enūma Anu Enlil* recently published by Gehlken 2012.


<sup>35</sup> +KBo 30.31+FHL 17+VSNF 12.53. Cf. Košak, hethiter.net/: hetkonk (v. 1.91), with further references.

<sup>37</sup> CTH 678.

<sup>40</sup> I have given more detailed indications concerning the presence of thunder festivals in local cult calendars and the period of their celebration in a communication presented at the 9th International Congress of Hittitology, held in Çorum (Turkey) on 1-7 September 2014.


Ünal A. 1998, *Hittite and Hurrian cuneiform tablets from Ortaköy (Çorum), central Turkey*, Istanbul.

Waal W. 2011, *They wrote on wood. The case for a hieroglyphic scribal tradition on wooden writing boards in Hittite Anatolia*, «Anatolian Studies», 61: 21-34.


# THE VENERATION OF LAMMA OF TAURISA AND THE DISCREPANCIES BETWEEN VERSIONS OF THE AN.TAḪ*.*ŠUM FESTIVAL

*Niccolò Galmarini*

#### *Abstract*

In the complex textual corpus of the AN.TAḪ.ŠUM-festival we can identify different versions which are often very dissimilar. This is demonstrated by the comparison of the outline tablets and the preserved daily descriptions. Sometimes, these discrepancies seem to be ascribable to different periods or to specific cultic necessities of various kings. In other cases, however, different versions of the festival seem to belong to alternative coexisting traditions. It is often very hard to define the diachronic relation between them. In the present paper, we examine the case of god LAMMA of Taurisa, whose veneration was not so relevant in the great spring festival, but it is very useful in distinguishing at least two different versions of the festival, represented in particular by the outline tablets A and G. It is also essential in order to classify synchronically and diachronically the daily descriptions of CTH 615 (days 22-25, Festival for Ištar of Ḫattarina) and CTH 617 (day 32, Festival for LAMMA of Taurisa). Two distinct traditions emerge: one, where the celebration of the deity is completely absent and to whom belong the outline tablet A and some Middle Hittite texts of CTH 615; the other, where the veneration of LAMMA of Taurisa was strictly connected with that of Ištar/Ša(w)uška of Ḫattarina and to whom belong the outline tablet G, other Middle Hittite fragments of CTH 615 and the texts of CTH 6171 .

The text of the great AN.TAḪ.ŠUM-festival has been transmitted through a long and rich textual tradition which crossed all the period of the Empire and to whom belong a large number of documents2 . These documents are often very hard to classify in order to reconstruct the festival program due to their fragmentary nature and the existence of different versions of the festival as clearly testified by the outline tablets.

\*\*\*

The aim of this contribution is to analyze a section of the AN.TAḪ.ŠUM-festival, where the discrepancies between the outline tablets and the daily descriptions are particularly evident and present to us some strong interpretative problems regarding the study of the great Hittite festival. These are the days dedicated to god LAMMA of Taurisa.

LAMMA of Taurisa rarely appears in the Hittite religious texts. It was part of the local pantheon of Taurisa, together with the Sun-God Tiwat and his paredra Kamrusepa3 and it often appears in the myths of Luwian origin4 . The god is present also in the AN.TAḪ.ŠUM-festival program but not in all versions.

#### *1. The outline tablets of the* AN.TAḪ.ŠUM*-festival (CTH 604)*

In the main outline tablet referred to the festival, A (KBo 10.20), dated to the time of Tutḫaliya IV, on which Güterbock's reconstruction is based5 , we cannot find the name of LAMMA of Taurisa. The


#### 50 *NICCOLÒ GALMARINI*

outline tablet A, in correspondence with the days where other outline tablets celebrate the god (approximately after day 20), dedicated six days to Ištar/Ša(w)uška of Ḫattarina, from day 22 to 27, after the celebrations in the temple of the Storm-God of Aleppo on day 21.The festival started at Ḫattusa in the temple of Aškašipa; during the following days other temples were visited, such as the one of Ištar/ Ša(w)uška of Ḫattarina (day 23) and the one of Ninurta (day 24). After day 25, when a lamentation for the goddess were recited, the last two days were dedicated respectively to the meat offerings to the goddess Ištar/Ša(w)uška of Ḫattarina, to the Ninattanni goddesses and to Karmaḫili (day 27)6 .

LAMMA of Taurisa is mentioned neither in the outline tablet A nor in the older duplicate B (KUB 30.39+ dated to the 14th century BC)7 where this part of the celebrations is not preserved, but it was venerated according to the other outline tablets of the festival.

The outline tablet G8 (VS NF 12.1) is also datable to king Tutḫaliya IV, but it is the only exemplar of a parallel coeval tradition which is different from that of tablet A and, according to Houwink ten Cate (2003: 205-219), it would not describe the rites in their definitive form. On day 24 the king celebrated Ištar of Ninive and another god. Next day, in the occasion of the rites for Ištar/Ša(w)uška of Ḫattarina, the king and the queen went to the 'garden of secrecy', where they celebrated LAMMA of Taurisa, another god and Ea. The god LAMMA of Taurisa was celebrated also on the following day (26) in the temple of the Storm-God of Aleppo9 .

<sup>6</sup> Rev.III

23' *lu-uk-kat-ti-ma* LUGAL MUNUS.LUGAL *I-NA* ⌈ É⌉ <sup>D</sup>*Aš-ka-ši-pa*

24' *pa-a-an-zi* LÚ.MEŠḪAL-*ma* <sup>D</sup>*IŠTAR* U[ RU*Ḫ*]*a-at-ta-ri-na*

25' [*mu-u-ga-an*]*-zi* UD.22.KAM

\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ 26' [*lu-uk-kat-ti-ma* LU]GAL MUNUS.LUGAL [*I-NA* É *IŠTAR* URU*Ḫ*]*a-*[*at-ta-r*]*i-*⌈ *na*⌉ 27' [*pa-a-an-zi* LÚ.MEŠḪAL-*ma*] ⌈ *A*⌉ *-NA* <sup>D</sup>*I*[ŠTARURU*Ḫa-at-t*]*a-ri-na*

28' [*mu-u-ga-an-z*]*i*

\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ 29' ⌈ *lu-uk-kat-ti-ma*⌉ LUGAL MU[NUS.LU]GAL *I-NA* É D[IB? ] *pa-a-an-zi*

30' ⌈ *nu* šu-up-pa *wa-ar-ap*⌉ *-pu-u-wa-an-zi* LÚ.MEŠḪAL-*ma*

31' <sup>D</sup>*IŠTAR* URU*Ḫa-at-ta-ri-na nam-ma mu-u-ga-an-*[*zi*]

32' *ŠA* DINGIR*LIM-ya* ŠU<sup>Ḫ</sup>I.A *šu-up-pí-ya-aḫ-ḫa-an-zi*

33' *ne-ku-uz me-ḫur-ma ku-lu-mur-ši-ya* ši-pa-an-da-[*an-zi*]

\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ 34' *lu-uk-kat-ti-ma A-NA* <sup>D</sup>*IŠTAR* URU*Ḫa-ad-da-ri-n*[*a mu-u-ga-an-zi*? ]

\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ 35' *lu-uk-kat-ti-ma A-NA* <sup>D</sup>*IŠTAR* URU*Ḫa-ad-da-ri-n*[*a mu-u-ga-an-zi*? ]

36' šu-up-pa-ya-ašUD-*az*

'Next day the king (and) the queen go to the temple of Aškašipa. The seers [invo]ke Ištar of [Ḫ]attarina.Day 22. [Next day the ki]ng (and) the queen [go to the temple of Ištar of Ḫatt]a[ri]na; [then, the seers invo]ke Iš[tar of Ḫatt] arina. Next day the king (and) the [que]en go to the temple of [Ninurta? ] for the meat washing, while the seers invo[ke] Ištar di Ḫattarina and purify the hands of the deity. In the evening they off[er] *kulumuršiya* offerings. Next day [they make an invocation? ] to Ištar di Ḫattarin[a]. Next day [they make an invocation? ] to Ištar di Ḫattarin[a]; it is the day (of) meat (offerings). Next day the seers make offerings to the Ninattanni goddesses, while the king celebrates Karmaḫili'.

<sup>7</sup> For the dating of the outline tablets A and B, see Houwink ten Cate 1986: 96, 106-110.

<sup>8</sup> F according to Houwink ten Cate (2003).

<sup>9</sup> Rev.

\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ 2' [*lu-uk-kat-ti-ma* LUGAL]-*uš I-NA* É-*TIM* GAL EZEN4 [AN.TAḪ.ŠUMSAR *A-NA* D...]

3' [Ù *A-NA* <sup>D</sup>*IŠTAR* UR] <sup>U</sup>*Ne-nu-wa* EZEN4 AN.TAḪ.ŠUMSAR x[ ] 4' [SISKUR *ku-lu-mur*]*-ši-ya* x [... UD.24.KAM]

\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ 5' [*lu-uk*]*-kat-ti-ma A-NA* DLAMMA URU*Ta-a-u-ri-iš-ša* [ D…] 6' *A-NA* DÉ.A-*ya I-NA* GIŠKIRI6 ḫ*ar-wa-ši-ya-aš* EZ[EN4 AN.TAḪ.ŠUMSAR *ŠA*? ] 7' <sup>D</sup>*IŠTAR* URU*Ḫa-at-ta-ri-na i-ya-zi* UD.25? .[KAM]

\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ 8' *lu-uk-kat-ti-ma* DLAMMA URU*Ta-a-ú-ri-ša* PÚAMA *K*[*a-li-im-ma* PÚ*Ka-li-im-ma* ] 9' *I-NA* É DU URU*Ḫal-pa-ya* SISKUR *ku-lu-mur-š*[*i-ya* UD.26.KAM]

\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ '[Next day the kin]g in the 'large building' the [AN.TAḪ.ŠUM] festival [for ... and for Ištar o]f Ninive the AN.TAḪ. ŠUM-festival ...[ *kulumur*]šiya [offerings] ...[... day 24]. [Next da]y for LAMMA of Taurisa [ ] and for Ea in the garden of secrecy (the king) celebrates the [AN.TAḪ.ŠUM] fe[stival of? ] Ištar of Ḫattarina. Day 25? []. Next day LAMMA of Taurisa, the spring Mother of K[alimma and the spring Kalimma] in the temple of the Storm-God of Aleppo *kulumurš*[*iya*] offerings*.* Day 26]'.

The so-called 'abridged version' D (KBo 45.16+)10 and its parallel E (KUB 44.39)11 partially correspond to the outline tablet A's program, but they do not mention Ištar/Ša(w)uška of Ḫattarina at all12. They dedicate, instead, specific parts of the rite to LAMMA of Taurisa and some deities belonging to its divine circle. On day 20 they celebrated, among the others, the spring Kalimma during a ḫ*adauri*  festival. Day 21 was dedicated to a ḫ*adauri* festival in the temple of the Storm-God of Aleppo, while on day 22 the king went to the temple of LAMMA of Taurisa.

Therefore, the situation described by the outline tablets is very complex and full of discrepancies. According to the outline tablet G, the rites for Ištar/Ša(w)uška of Ḫattarina were linked to those for LAMMA of Taurisa and Ea and they lasted only one day. Instead, according to A, the rites started in the temple of Askašipa and lasted five days with no celebrations for LAMMA of Taurisa. For the outline tablets D and E at one point the king went to the temple of LAMMA of Taurisa: they seem to correspond to G, but D and E do not mention Ištar/Ša(w)uška of Ḫattarina. The sequence of days in the outline tablets D and E correspond, however, to A's outline, although on day 22 they sensibly diverge. Days 25 to 27 in outline G seem to be parallel to days 20 to 22 in outline A and D-E, with the exceptions just mentioned. Outline A and G share again common features respectively from day 27 and 30 where god Karmaḫili is mentioned in both outline tablets (Mount Tapala is mentioned only in G; it will be celebrated next day according to outline A). On day 31 outline G mentions the forest of Aškašipa, where outline tablet A starts the celebrations for Ištar/Ša(w)uška of Ḫattarina on day 22.

#### *2. The daily descriptions*

#### *2.1. CTH 615*

The discrepancies found in the outline tablets are verifiable also among the daily descriptions. The analysis of the daily tablets of CTH 615 (festival for Ištar/Ša(w)uška of Ḫattarina) shows the existence of different drafts of the festival: at least two Middle Hittite versions which are datable approximately until the time of Suppiluliuma I; two Imperial versions (one in early New Script, the other in late New Script), both showing dependence on Middle Hittite originals13.


' …The king (and) the queen go in the 'large building'. [Nex]t day (20) the (holy) ablution (takes place). [ ] they celebrate Kalimma(naza). [In the] temple of Inar(a) they celebrate the *ḫadauri* festival, they slaughter [ten] sheep. [Nex]t day (21) the king [go]es to the temple of the Storm-God of Aleppo and to the temple of the Mother Goddess, for Kataḫḫa, Ugur, Ḫa[sa]mili and for Ea they celebrate the *ḫadauri* [festi]val, they off[er] ten sheep. [Next da]y (22) the king [go]es [to the] temple of LAMMA of Taurisa. [Wh]en? the house of the cedar in the forest…[ ]… at Ḫantilassaya …[ ] they cele[brate the Sun-goddess of ] Arinna'.

<sup>13</sup> MH/MS 1: IBoT 3.115+ (24th?, 25th, 26th days); KUB 45.37 (25th day); KUB 45.39 (26th day); KBo 30.61(+) (27th day); KUB 45.36 (day ?); KBo 51.128 (day ?); KBo 51.43 (day ?).

MH/MS 2: KUB 45.34+ (day ?), KUB 51.86 (day ?).

MH/MS 3?: KUB 47.70 (24th/25th day).

MH/ENS: KUB 41.13 (24th day); IBoT 4.28 (24th day); KBo 57.142 (24th day); KBo 57.143 (24th/25th day); KUB 45.38 (24th/25th day), KUB 47.70 (25th/24th); IBoT 2.55 (26th day); KUB 51.86 (day ?);

KBo 57.141 (day ?); KBo 48.121 (day ?); KUB 45.35 (day ?).

MH/LNS: KBo 46.102 (day ?); KBo 11.20 (day ?); VBoT 50 (day ?). For more details, see Galmarini 2013, 339.

#### 52 *NICCOLÒ GALMARINI*

Most Middle Hittite texts belong to what we call here version MH/MS 1, whose most ample manuscript is IBoT 3.115+ (CTH 615.1)14. The text surely belongs to the AN.TAḪ.ŠUM-festival for Ištar/ Ša(w)uška of Ḫattarina as testified by the worshipped gods (e.g., Ištar/Ša(w)uška of Ḫattarina and her hierodulae Ninatta and Kulitta) and the cultic operators (especially the LÚAZU). In the text we can identify various moments of the rite which are ascribable to day 25 of the outline tablet A, as demonstrated by Haas and Wegner (2005, 252-253): the text says that now (*kinun*) an invocation will be pronounced like on days 2 and 3 (Rev. IV 13'-16')15. Since we know from the outline tablet A that the festival started on day 22, that described here is day 25. I suggest that the preserved parts of IBot 3.115+ could describe also days 24 and 26 and the main elements which persuade me to refer IBoT 3.115+ to these specific days of the festival are:


Another quite ample fragment such as KUB 45.34+ (CTH 615.10)18, together with the LNS duplicate KBo 51.86 (CTH 470.1747)19, presents a completely different version of the rites. There, the place of celebrations for Ištar/Ša(w)uška of Ḫattarina and her hierodulae was the forest of Taurisa, which is completely absent from IBoT 3.115+ and the other fragments belonging to the version called here MH/MS 120.

The presence of the forest not only distinguishes KUB 45.34+ from the MH/MS 1 version, but it underlines the relationship between the text and the outline G, where the 'garden of secrecy' is mentioned during the celebration for Ištar/Ša(w)uška of Ḫattina and also with outline D, where a 'forest' is mentioned during the celebrations in the temple of Taurisa. Houwink ten Cate (2003, 213) suggested that the 'garden of secrecy' could be situated at the temple of Taurisa at Ḫattusa. I would suggest that the garden is in connection, or it is even to be identified, with the 'forest of Taurisa', which occurs in KUB 45.34+, and in some LNS daily tablets classified under CTH 617 (AN.TAḪ.ŠUM-festival for LAMMA of Taurisa).

<sup>14</sup> For the full text, see Wegner 1995: 111-116 (nr. 19); Groddek 2012: 72-74.

<sup>15</sup> Rev. IV

13' [*I-NA*] UD 2.KAM-*ma na*[*m-ma* LÚAZU *ma-aḫ-ḫa-an mu-ki-iš-ki-it*]

14' [*ki-nu-u*]*n*? *QA-TAM-MA mu-g*[*a-a-iz-zi* ]

15' [*I-NA* U]D 3.KAM-*ma nam-ma* LÚA[ZU *ma-aḫ-ḫa-an mu-ku-iš-ki-it ki-nu-un*]

16' [*QA-TAM-M*]*A mu-ga-a-iz-zi* LÚ.MEŠ[ ]

'Then, on the second day [no]w? he make[s an invocation] in the same way [as the seer makes an invocation]. Then, [on the] third day [now] he makes an invocation [in the same w]ay [as] the s[eer makes an invocation], the men[ ]'.

<sup>16</sup> Obv. I

20' *nu* GÌRMEŠ-*aš ku-it wa-a-tar ki-iš-ša-*[*r*]*a-aš* KÙ.BABBAR

21' *A-NA* DINGIR*LIM* GÙB-*la-az ki-it-t*[*a-r*]*i* ]

22' *PA-NI* DINGIR*LIM* GIŠ*pu-u-ri-az da-a-*⌈*i*⌉[]

23' *ke-e-ez ke-e-ez-zi-ya d*[*a-a-i* ]

24' ša-ra-a7-ŠU *kar*! -*ap-z*[*i nu* ŠADINGIR*LIM* GÌRMEŠ-*aš*]

25' *ar-ru-ma-aš* SÌR SÌR[*RU* ]

'And that water of the feet which stays in a silver hand-shaped vessel at the left of the deity, [ ] (he) places it in front of the deity with a stand *puri-* [ ] he pl[aces (it)] here and there, he lifts (it) seven times up [and] they si[ng] the song of washing [the feet of the deity]'.

<sup>17</sup> Rev. IV

25' [ ]x[ UZU]NÍ[G.GIG UZUŠÀ IZ]I-⌈ *it*⌉  *za-nu-an-zi*

They roast [l]ive[r and heart with fi]re.


7' [ *a-ra-a*]*ḫ-za A-NA* GIŠTIR URU*T*[*a*]*-ú-ri-ša-kánan* [ ]

8' [ ]-*zi nu* LÚAZU *pa-aḫ-ḫ*[*ur da-a-i nu*] *wa-ar-nu-zi nu* 3 MUŠEN.[GAL? ]

9' [ *A-NA* <sup>D</sup>*IŠT*]*AR* URU*Ḫa-at-ta-ri-na* <sup>D</sup>*Ni-na-at-t*[*a*]

10' [ <sup>D</sup>*Ku-li*]-*it-ta am-ba-aš-ši wa-ar-nu- zi /*

'[ ou]t of the forest of T[a]urisa []…and the seer [takes] the fi[re and] burns, then three gee[se? to Išt]ar of Ḫattarina, Ninatt[a (and) Kuli]tta he burns the *ambašši* offering'.

#### *2.2. CTH 617 and KUB 2.8*

Among the texts of CTH 617, KUB 2.8 is the most ample exemplar21. There the forest of Taurisa constitutes the main set for the rites to the deity as demonstrated by the colophon22.

The tablet contains the best preserved section of the rites for LAMMA of Taurisa during the AN.TAḪ. ŠUM-festival in the forest of Taurisa and in a temple dedicated to the deity. The text is a copy datable to the end of the Empire and it testifies a late version of the festival program.

The beginning of the rite is full of gaps and its structure is difficult to reconstruct. It describes the arrival of the king(?) (*a-ri*) (Obv. I 8') and his entry in the temple (*a*]*n-da pa-iz-zi*) (Obv. I 10') to start the celebrations. The following rites are quite standard cultic operations but with some peculiarities. One of the main features of the rite is the veneration of a sequence of deities and divine entities repeated many times in the text with some variations, such as LAMMA of Taurisa, the springs Kalimma and the Mother of Kalimma, Ḫasamili, Assiyaza ('the beloved'), the Storm-God of the Forest, Zuliya, LAMMA of the river, gods Salawani of the door23. These deities were celebrated individually or in groups.

In Rev. IV the second section of the rites starts. It seems to be more characterized by the action of the king who drinks in honor of the same sequence of deities. It offers sometimes 'inside' (*andurza*), sometimes 'outside' (*ašgaza*), probably references to a temple building. This section continues also in Rev V.

KUB 2.8, which is quite late from the paleographic point of view24, shows similarities to:


KUB 2.8 shows features in common with both outline tablets G and D which follow different sequence of days in the festival celebration. Instead, it has nothing in common with outline tablet A's version. Güterbock (1960, 98) suggested that the texts of CTH 617 could belong to day 32 of the outline tablet A, but his suggestion is not supported by concrete evidence: in fact, on day 32 the king went to Ḫaitta and started the celebration for Mount Puskurunuwa25.

According to Forlanini (2008: 169), the city of Taurisa could be situated in the area of Ištaḫara, or in the south-east given the connection with the river Zuliya and Mount Daḫa, but it is hard to believe that, at least at the end of the Empire, the king could leave the capital and go to Taurisa at that point of the festival (according to outline tablet G on day 26 they performed the rite in the temple of the Storm-God of Aleppo at Ḫattusa)*.* We cannot exclude that originally the itinerary could include a journey to Taurisa, as suggested by the veneration of the springs Kalimma and Mother of Kalimma, which probably had a real geographical position (Taurisa was actually visited during a cultic journey for the

2 [*iš-tar-n*]*i* -*ya-aš* EGIR-*an tar-nu-um-ma-aš*

3 [ŠA] DLAMMA URU*Ta-ú-ri-ša* \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_

'[Tablet x] finished, [(that) of the middle o]ne of leaving behind [of] LAMMA of Taurisa'.

When the king celebrates the AN.TAḪ.ŠUM-[fest]ival in the forest of Taurisa for LAMMA of Taurisa. It is true to a [writing board].

<sup>23</sup> They were celebrated in group in Obv. I 19'-24', Obv. II 37-43, Rev. V 18'-24'. The spring Kuwanniya receives individual offerings in Obv. I 37', Obv. II 28, Obv. III 9, Rev. V 21'. 24 It is classified as sjh. according to the website hethiter.net/hetkonk (v. 1.87).

<sup>25</sup> See Galmarini 2014: 277-295.

<sup>21</sup> For the full text see Groddek 2009: 59-67. Other texts classified under CTH 617 are KBo 13.176, KBo 47.244, KUB 10.81, KBo 61.142, Bo 6271. These texts contain the names of the same deities known from KUB 2.8 in very fragmentary contexts and do not provide crucial information about the rite.

<sup>22</sup> Rev. VI

<sup>1</sup> [DUB.x.KA] M *QA-TI* !

<sup>4</sup> *ma-a-an-za* LUGAL-*uš I-NA* GIŠTIR URU*Ta-ú-ri-ša*

<sup>5</sup> *A-NA* DLAMMA URU*Ta-ú-ri-ša*

<sup>6</sup> [EZE]N4 AN.TAḪ.ŠUMSAR *i-ya-zi*

<sup>7</sup> [*A-NA* GIŠ.ḪUR-*k*]*án ḫa-an-ta-a-an*

#### 54 *NICCOLÒ GALMARINI*

celebration of some LAMMAs in KUB 9.1726). But soon, due to the necessity of saving time in the extremely rich festival program, probably the king started to celebrate the deity in its temple at Ḫattusa, as clearly told by the abridged version outline tablet D. Consequently, the forest of Taurisa would be another designation for the 'forest' of the outline tablet D and the 'garden of secrecy' of the outline G, or maybe they could be included in the same cultic space.

#### *3. Final remarks*

We have to consider the existence of another version of the festival, as old as the main version testified by the outline tablets A and B − according to the paleographic analysis − whose tradition survived until the end of the Empire (given the existence of LNS duplicates). There, the veneration of Ištar/Ša(w) uška of Ḫattarina was connected to the temple of LAMMA of Taurisa, rather than to that of Aškašipa, as in the main versions of the festival A. The Middle Hittite dating of KUB 45.34+ testifies that this version is more or less coeval to that of IBoT 3.115+. So, we can admit the existence of synchronic differences in the festival celebration, regarding in particular the sacred places involved. The confirmation of the existence of at least two versions is the fact that the two Middle Hittite daily versions are related respectively to the outline A and to the outline G. The position of the abridged versions D is still not clear to us, even though the presence of LAMMA of Taurisa suggests us to admit its relationship to the second version of the festival. As for the case of CTH 617, I think that those daily tablets are related to the second version, but also even if Güterbock's suggestion were correct and it would be considered part of outline tablet A's version (day 32, before the arrival of the king at Ḫaitta, and I think that there was not enough time to stop also at Taurisa), the existence of two different versions of the festival would not be in doubt: the relationship between Ištar/Ša(w)uška of Ḫattarina and LAMMA of Taurisa would be in any case a strong parameter to distinguish one version to the other: in version A and MH/MS 1 this relationship is absent.


It is not easy to explain through historical contingencies or the action of specific kings the existence of different versions of the festival. Considering Houwink ten Cate's dating of the outline tablets, we suggest that version of A and IBoT 3.115+, that I called here version MH/MS 1, reflected the image of the festival after the cultic reorganization of Tutḫaliya IV. It had, probably, larger diffusion than the others, as shown by the number of preserved texts which follow this version. On the contrary, I suggest that the version of G and KUB 45.34+ (MH/MS 2), and probably that of the texts mentioning LAMMA of Taurisa such as CTH 617, although as old as the first one and survived until the time of Tutḫaliya IV, could have been penalized by the last standardization operated by the king.

The presence of LAMMA of Taurisa and its relationship to the cult of Ištar/Ša(w)uška of Ḫattarina is a valid criterion to classify the textual material related to the AN.TAḪ.ŠUM-festival. I have attempted to show the existence of different versions of the festival where the itinerary and the celebrated deities were quite dissimilar. These are two coeval versions, as demonstrated by the existence of Middle Hittite copies in both cases, whose textual tradition arrived till the time of Tutḫaliya IV. One seems to have prevailed on the other, as shown by the largest number of related fragments, even though it is hard to define their real fortune in the cultic practice. We can suppose that such different version of the rites responded to cultic needs of different periods and kings. However, in this case we cannot identify the historical contingencies which brought to the existence of two different versions. Nevertheless, we must admit that the textual transmission of each version was guaranteed during all the Empire period, even if just in copies of archive and not in the effective cultic practice. This is one of the opened questions that we have to consider carefully when we study the AN.TAḪ.ŠUM-festival and its textual tradition.

<sup>26</sup> CTH 685, text in McMahon 1991: 218-221.

#### *Bibliography*


# LOOKING FOR ZIPLANDA. THE HITTITE NAMES OF KUŞSARAY AND KALETEPE

*Piotr Taracha*

#### *Abstract*

Two different localizations of Ziplanda, the Hittite sacred city, and the Daḫa mountain connected with it have been proposed in research to date. The one identifying Ziplanda with Uşaklı/Kuşaklı Höyük (Yozgat) is based on an idea from the 1930s linking Ankuwa from Hittite sources, said to lie near Ziplanda, with Amkuwa (modern Alişar Höyük) known from Cappadocian tablets; if so, then Daḫa would be the modern Kerkenes Dağ. The other one challenges the identification of Ankuwa with Amkuwa-Alişar, suggesting instead that Ziplanda be identified with Alacahöyük and the Daḫa mountain with Kalehisar, this in view of the fact that Hittite texts point to a more likely location of Ankuwa to the north or northeast of Ḫattusa. The present paper argues in favor of a new identification, linking Ziplanda with Kuşsaray, located c. 15 km to the northeast of Çorum, and Daḫa with the nearby Kaletepe. The new idea is fueled by a fragmentary text with a description of the spring festival in Ziplanda, found among the tablets from Kuşsaray. Topographical data on Ziplanda from Hittite sources has also been compared with the situation on the ground today. The landscape of Kuşsaray and Kaletepe appears to argue in favor of the presented hypothesis identifying the site with the Hittite Ziplanda.

Progress in studies on Hittite geography has been significant recently (e.g., Kryszeń 2014; Strobel 2008). Of the three sacred cities of the land of Ḫatti, Nerik, Arinna and Ziplanda, the first has been located at Oymaağaç Höyük and confirmed by new excavations carried out by a team from the Freie Universität in Berlin (Czichon 2009; 2010; Czichon, Klinger 2005; Czichon *et al*. 2011; cf. Klinger 2009), which have uncovered fragmentary tablets twice mentioning the *daḫanga*-, a cultic structure linked with the temple of the Storm-God of Nerik (Lamante 2014), as well as the deity Daḫangaili1 . Arinna is in all likelihood to be identified with Alacahöyük (fig. 1) (Erkut 1992; now also Taracha 2011; 2012), the identification being argued for still further by a Middle Hittite tablet Esy 11/20 from Eskiyapar, discovered in 2011, containing a festival description (Sipahi 2012: 50, 60 fig. 9) 2. The text speaks of people from Arinna and the city of Taḫurpa, which is known to have lain close to Arinna, and it also mentions the Kaskaeans. In the light of this text, the location of Taḫurpa at Eskiyapar, northeast of Ḫattusa, seems likely indeed (Sir Gavaz 2012: 34-37)3 . Following an independent analysis of Taḫurpa's regional cluster based on Hittite textual evidence, A. Kryszeń concludes:

\*\*\*

Perhaps the reason for Taḫurpa's political and administrative insignificance, which emerges from the sources, was its position between two powerful centres, Ḫattusa and Katapa, both within a distance of a day's journey, as clearly shown by the AN.TAḪ.ŠUM and *nuntarriyašḫaš* tablets.

<sup>1</sup> Pers. comm. J. Klinger. See also R. Czichon's paper *Oymaağaç Höyük Kazı Sonuçları 2014*, read at the ninth International Congress of Hittitology in Çorum, 1-5 September 2014. 2 See also A. Süel's paper *Eskiyapar tableti*, read at the ninth International Congress of Hittitology in Çorum,

<sup>1-5</sup> September 2014. 3 *Contra* S. de Martino (de Martino, Fales, Ponchia 2010: 194-196), who would like to place Taḫurpa at

Yassıhöyük (Yozgat), southeast of Ḫattusa.

It is also quite possible that the KASKAL GAL, 'the main road', which started in Ḫattusa, led through Taḫurpa to Katapa.

The distance between Taḫurpa and both Ḫattusa and Arinna must have been insignificant, as demonstrated by the fifth day of the *nuntarriyašḫaš* festival. The road from Taḫurpa to the capital was covered on foot as well as by a *ḫuluganni*-cart, whereas the distance between Taḫurpa and Arinna could have been covered twice on the same day.

Also one day separated Taḫurpa from Ḫanḫana, although this time the detailed itinerary is unfortunately not known […].

What concerns directions, the fact that Ḫanḫana served as a base for military campaigns against the Kaškaeans would necessarily pull Taḫurpa roughly to the north of Ḫattusa, or at least render impossible its location south of the capital4 . Katapa, on the other hand, shows no direct connection with Ḫattusa, but strong links to Zippalanda and indirectly to Sapinuwa (modern Ortaköy). Taḫurpa can be therefore tentatively placed north by northeast of the capital (Kryszeń 2014: 120).

The identification and localization of the third of the sacred cities, Ziplanda, continues to be debated, with two competing hypotheses in the literature, which the present article adds to by arguing in favor of an entirely new localization of this Hittite centre. It is proposed by the present author to see Ziplanda in Kuşsaray located about 15 km northeast of Çorum (fig. 1)5 and in consequence to identify the mountain of Kaletepe, rising about 500 m to the southwest of the modern village, with Daḫa known from Hittite sources.

O.R. Gurney (1995) first identified ancient Ziplanda and the associated Mount Daḫa with Uşaklı/ Kuşaklı Höyük (Yozgat) and the granitic Kerkenes Dağ massif just about 8 km to the south. His idea, based on the identification of Ankuwa with Amkuwa (modern Alişar Höyük) of the Cappadocian tablets proposed by I.J. Gelb (1935: 9-10), has been widely accepted (e.g., Crasso 2005; 2008; Gorny 19976 ; Pecchioli 1999; Forlanini 2002: 261 with n. 18; 2008: 155-156). It has also served as a working theory for the Italian mission of the University of Florence excavating currently at Uşaklı/ Kuşaklı (Pecchioli Daddi, Torri, Corti 2014: 672-673). As a matter of fact, however, the evidence of Hittite texts in favor of this identification of Ankuwa (= Hattian Ḫannikku, cf. Goetze 1962: 29; Laroche 1962: 29) with Amkuwa-Alişar is not unambiguous. On the contrary, the role of Ankuwa in the organization and supply of the cult to Telipinu in Ḫanḫana (Haas, Jakob-Rost 1984: 40, 44, 47), a town lying most probably in the vicinity of modern Çorum or somewhat to the north of it (Camatta 2006), indicates that it, Ankuwa, was also north(east) of Ḫattusa (Popko 1995: 256-257), although likely farther to the north than suggested by M. Popko, who identified Ankuwa with Eskiyapar (Popko 2000: 447). Admittedly, the road between Ziplanda and Ankuwa could not have been long, if it were covered twice during the same day. During the spring festival in Ziplanda and on Mount Daḫa, the king completed the ceremonies on the mountain, got into his chariot and went to Ankuwa. The ceremonies continued in his absence until at the end the Storm-God of Ziplanda returned 'to his place' in the temple, after which a messenger was sent to Ankuwa to inform the king that the festival had been successfully completed (KUB 20.96 iii 19'-iv 22, v 5-11, Popko 1994: 192- 197; Kryszeń 2014: 315-316). The above indicates that Ziplanda's location should also be searched for north or northeast of Ḫattusa. M. Popko (1994: 13, 29-31, 90-91; 2000) proposed to identify Ziplanda with Alacahöyük and Mount Daḫa with Kalehisar, about 4 km north of the site. However, as noted by the present author elsewhere: «Since neighbouring Katapa must be sought now in the direction of the Zuliya river (modern Çekerek), probably to the north or northeast of Sapinuwa/Ortaköy, Ziplanda is likely to have been situated rather farther north or northeast than suggested by Popko» (Taracha 2010: 351; cf. also 2009: 24 n. 108)

At the time the actual location of Ziplanda was still not clear to my mind. This changed with the fortunate publication by R. Akdoğan (2010: 9-10, 69-70, 173-174; 2011: nos 17, 143, 388-390) of five tablet fragments from Kuşsaray, now held in the Museum of Anatolian Civilizations in Ankara. O. Soysal (2011: 7, 26) recognized one of the fragments, ABoT 2.143, as belonging to the description of the spring festival in Ziplanda (CTH 635). The preserved text parallels KUB 11.30 + KUB 44.14 iv 21'-27' (Popko 1994: 210-211). ABoT 2.143 reads (Akdoğan 2010: 69-70, without restorations):

<sup>4</sup> *Contra* Forlanini 2008: 155.

<sup>5</sup> The suggestion was first presented by the present author in 2012 without wider comment: Taracha 2012: 109.

<sup>6</sup> Later, however, Gorny (2006: 30) identified Ziplanda with Çadır Höyük near Alişar, which he was excavating.


«(The king goes into the *huwaši*-stone-sanctuary of Mt. Daḫa. The king bows to the *ḫuwaši*-stone.) The king (gives) one silver disk of five shekels, one ox (and) one sheep as a gift. He bows to the deity7 . // The king steps (and) breaks nine *dannaš*-breads: Four *dannaš*-breads he breaks for the *ḫuwaši*-stone of Mt. Daḫa, two *dannaš*-breads he breaks for the hearth, one *dannaš*-bread he breaks for the spring, whereas from two *dannaš*-breads one makes *ḫarzazu*-breads and the *ḫam*[*ina*]-priest puts them on a wooden plate».

In the present author's opinion, the text comes down strongly in favor of the identification of Kuşsaray with the Hittite Ziplanda. Where else, beside the capital city, would a text describing a local festival in Ziplanda be held, if not in Ziplanda itself? A similar case can be argued with regard to the spring festival of the Storm-God of Sarissa (CTH 636), descriptions of which were stored both in Ḫattusa and in the archive in building A at Kuşaklı-Sarissa (Wilhelm 1997: 9-14, 17-19).

Another text from Kuşsaray, ABoT 2.17, a fragment of the Ḫedammu Myth (CTH 348.27.B), had been known earlier in H. G. Güterbock's transcription, published as KBo 26.109, where 'Çorum' was given as its provenience8 . The text parallels KBo 26.79 (Groddek 1998: 231-232, no. 44). O. Soysal noted (2011: 20) that «[i]t is remarkable that a composition of this kind (a myth of Hurrian origin) was really unearthed outside of Boğazköy». The circumstances of making the transcription were given by Güterbock as follows: «Von mir 1939 im Amtszimmer des Unterrichtsdirektors der Provinz Çorum transkribiert. Von dort muß das Bruchstück mit anderen in das Lokalmuseum von Hüyük gebracht worden sein, wo ich es aber später nicht mehr sah» (Güterbock *apud* Güterbock, Carter 1978: VII).

This information illuminates to an extent what happened later to the tablets from Kuşsaray. Peasants had found them on the local höyük in the late 1930s and subsequently, the tablets were transferred to the Provincial Directorate of National Education in Çorum, where H.G. Güterbock had the opportunity to see them before they were sent to the museum. His assumption that they were later transferred to the museum in Alacahöyük appears to be confirmed by information from the museum director there that its collection still holds two tablets marked as originating from Kuşsaray9 . The two tablets, which are not noted in S. Košak's *Konkordanz*, have yet to be located in the local museum.

The five texts that were transferred to the museum in Ankara were recorded in the museum register in 1976, suggesting that that was when the transfer occurred. The remaining three texts from this set, preserved in extremely fragmentary condition, can be described only tentatively: ABoT 2.388 oracular letter(?); ABoT 2.389 festival fragment(?); ABoT 2.390 letter(?). The inventory numbers of the five tablets from Kuşsaray, from 2 to 6, indicate that at least one other find was transferred at the time to the museum in Ankara. It may have been a fully preserved Hittite vase, 40 cm high, which is known to have been found by the peasants of Kuşsaray 'on the northern slope of the höyük'. A brief note and a small and poor photo of this vessel (from the display of the museum in Alacahöyük?) was published in H.Z. Koşay's catalogue (1966: 91, 95 Fig. 9) where the vessel appeared as number one. The assumption that this vase was also transferred to Ankara has yet to be verified at the museum there.

In 1966, test excavations were carried out at the Kuşsaray Höyük:

On the orders of the Ministry of Education from May 24, 1966, on behalf of the Ministry and the Turkish Historical Society, the Alacahöyük excavation team conducted a sounding and survey between July 18-25, 1966, with fourteen workers at the Kuşsaray Höyük, 15 km east of Çorum, where reportedly some Hittite tablet fragments have been discovered (Koşay 1966: 89 [Turkish]; English transl. Soysal 2011: 26 no. 143, with emendations).

<sup>7</sup> For an English translation of KUB 11.30 + KUB 44.14 iv 19'–22', see CHD Š/3, 458.

<sup>8</sup> For ABoT 2.17: 3', see Rieken *et al*. 2009: note 2: ]*me-mi-aš-ma-wa*[; Akdoğan 2010: 9: -]x MI-*aš-ma-ši*[. More likely, however, -]x GE6 -*aš-ma-ši*[.

<sup>9</sup> Pers. com. A. Süel, whom the author would like to thank for facilitating contacts with the museum in Alacahöyük.

#### 60 *PIOTR TARACHA*

The results of this work were reported in brief by H. Z. Koşay (1966): The höyük measuring 150 x 100 m lies 300-400 m to the south of the Çorum-Samsun road, on a limestone hill rising naturally about 20-25 m. Two test trenches were opened (B and C), each 6 m by 6 m in area; both were excavated to a depth of 4 m, identifying three occupation phases. Stone foundations from Hittite times, on the surface, were damaged extensively by agricultural plowing as well as collecting stone building material for use in the village in houses and road construction. The finds were quite modest: two pieces of zoomorphic terracotta figurines, a cylindrical bead of a red stone with white veining, fragment of a stone vessel. Layers from the Early Bronze Age and the Chalcolithic were in better condition, yielding EB pottery comparable to the finds from Ahlatlibel and Alacahöyük, and Chalcolithic pottery (white on black) like the ceramics from Mersin XII, Kusura A, Büyük Güllücek and Beycesultan. There can be no doubt as to the great antiquity of this particular settlement center with roots at least in the fourth millennium BC. The association with Alacahöyük (Arinna?) is irresistible. In the third millennium BC already, before the growth of the importance of nearby Ḫattusa, Alacahöyük had been the seat of a local dynasty. It peaked in prosperity in the EB III period, which is how the royal tombs there are dated. One is entitled to expect an equally long history of the second of the Hittite sacred cities, that is, Ziplanda.

The hill Kaletepe that rises to the south of the village of Kuşsaray owes its name to a fort from Greek and Roman times located on its summit. Apart from finds from the period when it was a Greek and Roman defense point, H. Z. Koşay (1966: 89, 90) also noted a small stone vessel found earlier at the site, decorated with a bull's head in relief and sent to the museum in Çorum. No further information is available of this vessel and its dating.

In September 2013 the present author had the opportunity to visit Kuşsaray for a day to verify the topographical information in H.Z. Koşay's report. Having assumed that Kuşsaray be ancient Ziplanda, the author intended also to compare the situation on the ground with the data on the topography of Ziplanda from Hittite sources so masterfully put together and interpreted by M. Popko in his monograph of the city. In other words, the objective was to find out whether and to what extent Hittite texts confirmed (or excluded) the assumed identification.

Summing up conclusions reached by M. Popko (1994: 18-26), one can say that Ziplanda was a small town with its own defenses10. The main building there was a temple of the local Storm-God, presumably with a larger number of courtyards which housed the shrines of other deities (the most important one being a sanctuary of the Sun-Goddess of the Earth). The city also had its *ḫalentuwa*-building, where the king went to rest after coming to the city and where he prepared for the ceremonies to follow. He also slept there. The building may have been part of the Storm-God's temple complex or a local palatial complex (Taracha *forthcoming*). Upon arriving in the city the king descended from his chariot at the getting off place (*katapuzna*-) and went to a bathroom (É*taštappa*-), presumably before passing through the gate of the lower(?) town. Then he got on a *ḫuluganni*-cart to visit several cult places, after which he went through a gate to the *ḫalentuwa*-building.

According to the texts, Mount Daḫa was evidently near the town. During the ceremonies of the AN.TAḪ.ŠUM festival the king walked on foot from the temple of the Storm-God to the city gate and got into the *ḫuluganni*-cart already outside it. He then went first to visit the threshing floor (KISLAḪ), where he toasted the agriculture deities Šepuru/Zuwuru and Telipinu, and after that to the *ḫuwaši*stone-sanctuary of Mt. Daḫa (Bo 5110 + KBo 45.146: 1'ff., Popko 1994: 220-223; Roszkowska-Mutschler 2005: 195). On another occasion, during the spring and autumn festivals, the king was accompanied by the queen in a ceremonial passage to Mt. Daḫa (KBo 11.49 vi 1'-3', Popko 1994: 178-179). It should be noted that the king is progressing all the time in a slow ox-drawn *ḫuluganni*-cart, whereas during a similar ceremony in Sarissa (Kuşaklı), the vehicle of choice for the king to cover a distance of about 2.5 km from the temple of the local Storm-God to his *ḫuwaši*-stone-sanctuary outside the city was a chariot (Wilhelm 1997: 14; and in this volume). It could indicate that the road from Ziplanda to Mt. Daḫa was indeed much shorter.

We are also informed that the city was well visible from the *ḫuwaši*-stone-sanctuary (at the foot?) of Mt. Daḫa (Popko 1994: 26-29). Upon departing from Ziplanda the king visited once again the sanctuary on Mt. Daḫa, bowed to the city (KBo 13.214 iv? 9'-10'), then rode in a chariot either to Ankuwa (during the spring festival) or to Katapa (during the autumn festival) (KBo 13.214 iv? 13'-19', Popko 1994: 184-185). The sanctuary had its own gate (the texts speak of the gate of Mt. Daḫa). It appears that the Mountain-God was worshiped there together with the spring-goddess Anzili (*IŠTAR-li*). A sacred

<sup>10</sup> References to the relevant Hittite texts have not been cited here, they being found easily in Popko 1994.

pond (*luli*-), which is known to have lain outside the city (KBo 17.100 i 4, 12, Popko 1994: 138-141), may have also been connected with this sanctuary.

The höyük today stands on the northeastern fringes of the modern village of Kuşsaray, which lines a road passing north-south from the intersection with the Çorum-Samsun speedway (fig. 2). The houses occupy a natural dip in the land at the foot of Mt. Kaletepe and extend the full length of the eastern slope of the hill. By the same, the village occupies much of the area between the höyük and Mt. Kaletepe. Compared to H. Z. Koşay's description of the 1960s, the surface of the höyük is much more disturbed. It is practically impossible to discern any remains of walls on the ground. The distance as a crow flies between the höyük and the base of Mt. Kaletepe is no more than approximately 600 m. The ground dips slightly, then rises a little in the direction of the hill (fig. 3). A spring flows at the base of the hill on its northern side. Like many similar springs in central Turkey today, it has been walled in and turned into a watering place for animals (fig. 4). The hydrographic conditions in the area seem to have suffered compared to antiquity. A natural cut in the northern slope above the spring, filled with trees and bushes that are proof of greater humidity in the area, could indicate that what is today a channeled spring had actually flowed once as a stream. Should one look for the place of a sacred pond, which is known from Hittite sources, then the only possibility are the flat fields lying northwards a little lower down, between Kaletepe and the Çorum-Samsun road (fig. 5).

A dirt road runs from the village to the spring, then turns sharply and rises along the northern slope of the hill to about a half its height. This is the only access and it is relatively gentle. The eastern and southern rocky slopes are fairly steep; the western slope is severely eroded today, hence it is impossible to tell what it may have looked like in antiquity. From the place where the road ends there is an impressive view to the northeast to the village and höyük; the speedway to Samsun can be seen in the distance and beyond it a vast plain which was the hinterland of the ancient settlement on the höyük (figs. 2 and 6). From this point the easiest access to the summit of Kaletepe is from the east. The slope is covered with potsherds from different periods (fig. 7). On the summit itself, the bottom parts of the east and south walls of the ancient fort are well preserved. Fragmentary roof tiles from this period can also be seen on the ground.

Summing up, the landscape of Kuşsaray and the topographical relation between the höyük and Mt. Kaletepe do not exclude an identification with Hittite Ziplanda. Taking the point further still, it can be said that there is a good correlation between the situation today and data from Hittite texts. An intensive survey of the area and new excavations at the site could verify this idea.

Finally, one should recall the cursory information given by H. Z. Koşay regarding two settlement sites from the Hittite period located not far from Kuşsaray. The first one is Melikgazi situated 5 km south of Çorum, where Hittite pottery and polychrome Phrygian vessels have been found (Koşay 1966: 91). The other is even more interesting in the context of a discussion on the location of Ankuwa that, as was said, lay very close to Ziplanda. The site is Höyük, a mound measuring 100 m by 100 m, lying near the locality of Elvançelebi, about 7 km to the south of Kuşsaray, and well visible from a local road running from Çorum to Mecitözü. The surface of this mound was covered with pottery dated to the Early Bronze Age and to the Hittite period (Koşay 1966: 91).

#### *Bibliography*


Crasso D. 2005, *Ankuwa in Hittite Written Sources: Preliminary Observations*, «KASKAL», 2: 147-158.


#### 62 *PIOTR TARACHA*


Pecchioli F. 1999, *Review of Popko 1994*, «Bibliotheca Orientalis», 56: 151-157.


Sipahi T. 2012, *2011 Yılı Eskiyapar Kazısı*, in *2. Çorum Kazı ve Araştırmalar Sempozyumu*, Çorum: 45-61. Sir Gavaz Ö. 2012: *Involving the Whereabouts of the Taḫurpa City*, «KASKAL», 9: 31-43.


Taracha P. *forthcoming*, *Hittite Éḫalentuwa- Revisited*, «Altorientalische Forschungen».

Wilhelm G. 1997, *Kuşaklı – Sarissa. Band 1: Keilschrifttexte, Faszikel 1: Keilschrifttexte aus Gebäude A*, Rahden/Westf.

Figure 1. Map of Central-North Anatolia showing the sites mentioned in the text.

Figure 2. Kuşsaray and Höyük. View from Kaletepe.

Figure 3. Kaletepe. View from the village.

66 *PIOTR TARACHA*

Figure 4. The spring at the base of Kaletepe. Figure 5. The northern slope of Kaletepe (in the direction of the spring) and the fields between the hill and the Çorum-Samsun speedway.

Figure 6. Panorama of Kuşsaray. View from the eastern slope of Kaletepe.

Figure 7. Pottery fragments from Kaletepe.

# DIE AUSGRABUNGEN IN DER UNTERSTADT VON ḪATTUSA (2009-2014): ERSTE VORLÄUFIGE ERGEBNISSE

*Andreas Schachner* 

#### *Abstract*

Intensive surveys and geophysical prospection of the Lower City of Hattuša, started in 2009, enlightened the chronological and structural development of the city. Research focused in the area of Kesikkaya between the postern wall and the so-called Südareal of Temple 1. It allowed to draw conclusions about the overall continuous development of the area from the karum period until the 15th century BC in the Hittite period. This seems to suggest that unlike the written sources concerning the Anitta's destruction of the city, there was not a chronological hiatus between the two politically defined eras. Research in the area is indicative of the possible development of the entire city and its gradual transformation during the 17th century in a metropolis sustained by the ideological project of the emerging Hittite dynasty.

Durch die Forschungen in der Oberstadt von Ḫattusa wurde bis 2008 nicht nur deren strukturelle Entwicklung intensiv untersucht (Neve 1999, 2001; Schachner 2011a: 82-99; Schachner 2011b: 87- 93; Dürr 2014), sondern auch – nach einigen Wirrungen ‒ deren chronologische Entwicklung geklärt (Seeher 2006a, b). Vor diesem Hintergrund und auf der Grundlage vorbereitender geophysikalischer Prospektionen und Surveys werden seit 2009 intensive Forschungen in der Unterstadt mit dem Ziel durchgeführt1 , die chronologische und strukturelle Entwicklung der Altstadt im direkten Vergleich zur Oberstadt (Neustadt) darzustellen. Nach Abschluß der ersten, im Süden der Unterstadt lokalisierten Projektphase besteht die Möglichkeit, einige allgemeine Überlegungen darzustellen. Dabei ist beim derzeitigen Stand der Aufarbeitung deren vorläufigen Charakter zu betonen.

\*\*\*

Die Arbeiten konzentrierten sich im Süden der Unterstadt im Umfeld von Kesikkaya auf eine Terrasse zwischen der Poternenmauer und den bis 1970 ausgegrabenen Bereichen des Südareals (Abb. 1). Sowohl nördlich als auch unmittelbar südlich des bereits durch Surveys als bebaute Landmarke identifizierten Kesikkaya wurden durch Grabungen ganz unterschiedliche Aspekte der Siedlung geklärt, die in ihrer Kombination wesentliche Rückschlüsse auf die Gesamtentwicklung der Stadt während der frühen hethitischen Phase erlauben.

Ähnlich wie in den meisten Bereichen der Stadt nutzen die hethitischen Planer auch in der Unterstadt die verschiedenen Terrassen, um die Siedlung in funktional definierte Bereiche zu gliedern. Nördlich von Kesikkaya wurde auf einer der Terrassen, in die die Unterstadt unterteilt ist, trotz erheblicher Störungen durch nicht dokumentierte Suchschnitte vermutlich der Grabungen von Th. Makridi eine Abfolge von Architektur ergraben, die von der *kārum*-Zeit bis in das frühe 15. Jh. v. Chr. reicht (Abb. 2). Neben zahlreichen Einzelergebnissen und herausragenden Funden, die in den Vorberichten dargelegt wurden (Schachner 2010, 2011d, 2012a, 2014a), erlaubt die Erarbeitung einer durch 14C-Datierungen abgesicherten, architektonischen Abfolge zudem Aussagen zur allgemeinen Stadtgeschichte.

Im Norden des Areals wurde das Haus 90 ‒ ein hethitisches Hallenhaus ‒ praktisch vollständig freigelegt (Abb. 2). Obwohl keine datierbaren organischen Funde gemacht wurden, wird durch die statistische Auswertung der Keramik deutlich, daß dieses Bauwerk nur bis in das 15.Jh. v. Chr. genutzt wurde2 .

<sup>1</sup> Für die Vorarbeiten s. Dittmann, Röttger 2008: 133-136, 136-142; Dittmann, Röttger 2009: 46-53.

<sup>2</sup> Für die Grabungen: Schachner 2010: 167-168 Abb. 8; für die Datierung: Strupler 2013b: 166-170.

#### 68 *ANDREAS SCHACHNER*

Unter dem Haus 90 wurden im Verlauf der Grabungen mindestens zwei weitere Bauschichten erfaßt: die jüngere ‒ die in die älter-hethitische Zeit datiert ‒ stellt wahrscheinlich ein Gebäude des Zentralraumtyp dar (Abb. 3), der in Zentralanatolien seit dem 3. Jt. v. Chr. bis in die hethitische Großreichszeit in urbanen Zusammenhängen belegt ist (Schachner 2010: 167-168, Abb. 8; Schachner 2011a: 244-250, Abb. 119); von der darunterliegenden und aufgrund der Keramik datierten Schicht der *kārum*-Zeit wurden nur unzusammenhängende Mauern freigelegt, die aller Wahrscheinlichkeit nach auf dem gewachsenen Boden errichtet worden waren (Abb. 4)3 . Obwohl aufgrund der Störung der Befunde im mittleren Teil des Areals kein direkter Anschluß nach Westen und Süden möglich ist, kann diese Bauschicht allgemein als gleichzeitig mit dem *kārum*-zeitlichen Gebäude 82 angesehen werden, das ebenfalls auf dem gewachsenen Boden gegründet ist und das aufgrund von 14C-Datierungen in die zweite Hälfte des 18. Jh. v. Chr. datiert werden kann (Schachner 2011d: 40-41).

Die Grabungen im unmittelbar südlich anschließenden Bereich belegen, daß die ursprüngliche Topographie an dieser Stelle nach Süden abfiel. Denn unter dem Haus 91, aus dessen Fußboden ein 14C-Datum stammt, das diesen in das frühe 15.Jh. v. Chr datiert, und das einen stratigraphisch idealen *terminus ante quem* für die darunterliegenden Schichten darstellt (Schachner 2011a: 36 Abb. 9-10; Strupler 2013a: 159-164, Tab. 1, Nr. 10, Abb. 24), konnte eine 2,50 m tiefe Abfolge von Architektur, Schuttschichten und Fußböden bis auf den gewachsenen Boden ergraben werden (Abb. 5)4 .

Dabei vermittelt insbesondere die über die gesamte Abfolge gleichbleibende Ausrichtung der Architektur, die in allen Bauschichten keinen monumentalen Charakter hat, und die Tatsache, daß die Fundamente der jüngeren Bauten unmittelbar auf denen der älteren gegründet sind, den Eindruck einer von Beginn der Nutzung des Areals in der *kārum*-Zeit bis in das frühe 15. Jh. v. Chr. wahrscheinlich ungebrochenen Kontinuität (Abb. 2, 5). Gestützt wird dieses Ergebnis durch eine dichte Abfolge von Radiokarbondatierungen (Strupler 2013a: 164, Abb. 24-25)5 , die statistische und typologische Auswertung der Keramik und die Einheitlichkeit der botanischen Funde, die keine Brüche erkennen lassen (Pasternak 2012)6 . Anhand der Befunde gewinnt man den Eindruck einer organischen, ungebrochenen Entwicklung der materiellen Kultur in diesem nicht staatlich genutzten Teils der Siedlung. Offenbar findet der politische Umbruch, den die textlich überlieferte Etablierung der hethitischen Dynastie darstellt (Klengel 1999: 21-38), in den materiellen Befunden zumindest dieses Areals keinen erkennbaren Niederschlag.

Weitere 14C-Datierungen, die aus Tierknochen gewonnen wurden, die bei Ausgrabungen im Norden der westlichen Unterstadt in einem zwischen 1975-1977 ausgegrabenen Bereich gefunden wurden, deuten darauf hin, daß auch in anderen Bereichen der Unterstadt möglicherweise mit einer Siedlung während des 17. Jhs. v. Chr. zu rechnen ist (Tab. 1; Abb. 6)7 .


Tab. 1: Radiokarbondatierungen aus der nordwestlichen Unterstadt.

<sup>3</sup> Für eine Beschreibung der Funde aus diesem Areal vgl. Schachner 2014a: 93-95 Abb. 3-5.

<sup>4</sup> Für eine Beschreibung dieser Befunde: Schachner 2014a: 95-97.

<sup>5</sup> Weitere bereits vorliegende 14C-Daten verdichten den Befund und werden von N. Strupler in seiner Dissertation

(Univ. Straßburg) vorgelegt. 6 Eine detaillierte Darstellung der Befunde und Funde unter Verknüpfung mit den Radiokarbondatierungen ist durch N. Strupler als Teile einer umfassenden Arbeit zur Unterstadt von Hattusa in Arbeit. 7 S. a. Anm. 13.

In Ermangelung neuer Textquellen zu dieser Zeit kommt den archäologischen Befunden eine erhebliche Bedeutung für die Beurteilung dieser Übergangsphase zu. Die stratigraphischen Befunde sowie die 14C-Datierungen legen beim derzeitigen Wissenstand zumindest die Möglichkeit nahe, daß es entgegen des Berichts des Anitta-Textes und der auf diesem beruhenden, bisherigen Interpretation keinen hiatus in der Besiedlung zwischen den politisch definierten Epochen gab. Diese Besiedlung, deren Ausdehnung und Struktur noch weitgehend unklar ist, und die nicht unbedingt die Größe der *kārum*-zeitlichen Stadt gehabt haben muß, fiele dann in die Zeit der Labarna genannten Könige vor Hattusili I (Klengel 1999: 33-38).

Die beschriebene Entwicklung der Wohnbebauung in der südlichen Unterstadt kommt nach Ausweis der Befunde der Häuser 90 und 91 im frühen 15. Jh. v. Chr. zu einem etwa gleichzeitigen Ende8 . Weite Teile des Areals werden in der stratigraphischen Folge von einer Steinlage variierender Dicke überdeckt (Schachner 2011d: 34-35, Abb. 7-8), die entgegen früherer Vermutungen aufgrund der genannten Datierungsansätze bereits im Laufe des 15. Jhs. v. Chr. entstanden oder eingebracht worden sein muß. Die Bereiche südlich des Großen Tempels und des so genannten Südareals waren mithin wahrscheinlich seit dieser Zeit ähnlich wie die Ostseite der Gebäude nicht mehr bebaut. Wahrscheinlich sollte der Blick von der Stadtseite auf die spektakuläre Architektur bewußt frei gehalten werden (Abb. 7-8)9 .

Dieser Datierungsansatz wird durch architektur-typologische Überlegung zur Entstehungszeit des Großen Tempels (Müller-Karpe 2003), Grabungen in den Fundamenten des Südareals und der Tempelmagazine (Schachner 2011d: 31-34, Abb. 2-5; Schachner 2012a: 85-88, Abb. 2-4) sowie durch Beobachtungen bei Säuberungen im Zusammenhang einer detaillierten Bauaufnahme (2014) gestützt10. In allen Bereichen wurde ausschließlich Keramik gefunden, die aufgrund ihrer technischen und typologischen Merkmale der alt- oder allenfalls mittelhethitischen Epoche (parallel zu den Funden im Tal westl. vor Sarıkale) zugewiesen werden kann.

Als vorläufiges Ergebnis der Arbeiten in der südlichen Unterstadt, das durch weitere Arbeiten im Bereich des Südareals und des Tempels zu verifizieren ist, liegt die Vermutung einer allgemeinen Gleichzeitigkeit11 zwischen der Errichtung des Tempels und dem Ende der Nutzung des Areals zwischen der Monumentalbebauung und der Poternenmauer nahe. Dieser Zusammenhang deutet daraufhin, daß die Auflassung des zwischen 2009 und 2013 nördlich von Kesikkaya untersuchten Areals Teil einer größeren städtebaulichen Gesamtkonzeption an der Wende vom 16. zum 15. Jh. v. Chr. war, in deren Verlauf der Große Tempel, das Südareal und mit ihm eine völlig neue Ausrichtung der Unterstadt umgesetzt wurde.

*Südlich direkt anschließend an Kesikkaya*, dessen Integration in das städtebauliche Gesamtkonzept bereits durch Surveys der auf dem Felsen erhaltenen negativen Architektur belegt wurde (Dittmann, Röttger 2009: 48-53; Schachner 2011c), lag zwischen 2010 und 2014 ein zweiter Schwerpunkt der Arbeiten (Abb. 9). In diesem Areal beginnt die Nutzung erst in der hethitischen Epoche. Da am Fuß des Nordwesthanges von Büyükkale keine Bebauung vor der hethitischen Epoche nachgewiesen wurde (Schirmer 1969), ist z.Z. davon auszugehen, daß das Areal zwischen der Siedlung der *kārum*-Zeit in der Unterstadt im Westen und der Besiedlung auf dem Nordwesthang sowie auf Büyükkale im Osten wahrscheinlich unbebaut war. Die *kārum*-zeitliche Siedlung lag offenbar mit einem gewissen Abstand zur lokalen(?) Siedlung auf Büyükkale und dem Nordwesthang in der Ebene (siehe unten Abb. 16). Diese Freifläche, die etwa den Bereich zwischen Kesikkaya im Norden und dem Fuß des Nordwesthanges im Süden umfaßt haben muß12, wurde in alt-hethitischer Zeit mit öffentlichen Gebäuden ‒ dem Haus am Hang (Schirmer 1969: 18-27) und dem Gebäude vor Kesikkaya ‒ bebaut.

Durch die Grabungen zwischen 2010 und 2013 wurde ein großes, von Süden gegen den Felsen von Kesikkaya gebautes Gebäude der hethitischen Epoche freigelegt, das auf dem gewachsenen Bo-

<sup>8</sup> Im archäologischen Befund sind keine Spuren von Gewalt sichtbar, so daß eine bewußte Auflassung im

Rahmen einer umfassenderen Umgestaltung der Stadt wahrscheinlich ist. 9 Von Norden und Westen ist es nicht notwendig, den Großen Tempel durch eine offene Fläche in seiner Umgebung freizustellen, da er aufgrund seiner Gründung auf einer mehrere Meter hohen Steinterrasse aus diesen Richtungen von weither sichtbar war (Schachner 2014b; Schachner *im Druck* a). 10 Durch die Untersuchungen in 2014 wird erneut deutlich, daß im Großen Tempel, in den diesen umgebenden

Magazinen (vor allem im Süden) und im Bereich des so genannten Südareals während der früheren Grabungen nicht in jedem Raum auf den Fußboden gegraben wurde.

<sup>11</sup> In diesem Zusammenhang ist zu betonen, daß der Begriff 'Gleichzeitigkeit' im Zusammenhang bronzezeitlicher Kulturen und im Falle solch großer Bauvorhaben flexibel ‒ d.h. mindestens mit einem Spielraum von 2-3 Generationen ‒ verstanden werden muß. 12 Heute verläuft die moderne Besucherstraße durch dieses Areal.

#### 70 *ANDREAS SCHACHNER*

den errichtet wurde (Abb. 10). Trotz starker Zerstörungen durch die Grabungen 1911 von Th. Makridi (Schachner 2014a: 109-112, Abb. 28-35) kann man den Grundriß dessen östlichen und zentralen Teils rekonstruieren. Das Gebäude erstreckt sich auf zwei Ebenen in west-östlicher Ausrichtung vor dem Felsen. Auf der unteren, östlichen Stufe liegen die Räume 1 und 2, die von Osten zugänglich waren. Im Raum 3 hat sich im Norden vor dem Felsen eine Treppe befunden oder eine lange Rampe ging über diesen hinweg und führte den Weg in einen schmalen Korridor vor der Mauer 10-736. Der ursprüngliche Begehungshorizont von Raum 4 lag, wahrscheinlich auf einer Balkenkonstruktion, auf dem Niveau des Korridors vor der Mauer 10-736 und dem Fußboden von Raum 5. Durch den Raum 4 gelangte man wahrscheinlich an dessen südlichem Ende durch eine Tür, die in der Flucht mit den Türen der Räume 1 und 2 gelegen haben mag, in den Raum 5 / 6. Ursprünglich handelte es sich bei diesem um einen Raum, der zu einem unbestimmten Zeitpunkt durch den nachträglichen Einbau der Mauer 10-744 auf den Raum 5 reduziert wurde (Schachner 2014a: 103-109, Abb. 21-27).

Begrenzt wird der Raum im Westen durch eine ca. 25 m lange, nahezu vollständig erhaltene Reihe von unverzierten Kalkstein-Orthostaten (Abb. 10-12), die hangseitig vor eine mindestens 4 m tiefe, amorphe Struktur aus Lehmziegeln und Steinen gesetzt sind, deren rückwärtige Kante bisher nicht erfaßt wurde. Neben der Monumentalität des gesamten Gebäudes sind die Verwendung dieser aufwendig bearbeiteten Werksteine als echte Orthostaten sowie der ungewöhnliche Grundriß des Bereichs 5/6 die wichtigsten architektonischen Merkmale des Bauwerks. Dieser Raum hatte in seiner ursprünglichen Gestaltung die ungewöhnliche Form eines spitzwinkligen Dreiecks und ist in dieser Form ohne Parallele in der hethitischen Architektur. Ebenfalls ungewöhnlich ist der flachwinklige Knick am Übergang der Mauern 10-727 und 10-743, dessen Notwendigkeit beim gegenwärtigen Stand der Grabungen nicht erklärbar ist (Abb. 10).

In der Grundrißgestaltung fallen erhebliche Unterschiede zwischen der südlichen und der nördlichen Ecke des Bereichs 5/6 auf: während im Süden die Mauern aller Wahrscheinlichkeit nach in einem spitzen Winkel zusammentrafen, laufen sie in der nördlichen Ende trichterförmig aufeinander zu und sind mit einer Quermauer abgeschlossen (Abb. 10). Diese Ausgestaltung entspricht formal dem Grundriß der deutlich jüngeren Kammern 1 und 2 in der Oberstadt (s. unten Abb. 13) (Neve 1993: 72, Abb. 203.). Aufgrund der unterschiedlichen Gestaltung der beiden Ecken mag man vermuten, daß der nördliche Teil und die Mitte des Raums die Bereiche waren, in dem sich dessen eigentliche Nutzung abspielte, wobei im Moment noch unklar bleibt, welche Funktion ein hier in den Fußboden eingelassener Pithos hatte.

Ein markantes Element der mit Orthosthaten verzierten Mauer ist ein einzelner, großer, sehr gut gearbeiteter Steinblock, der etwa in der Mitte der Mauer angetroffen wurde (Abb. 11). Aufgrund der Fundlage ist klar, daß er Teil der Orthostatenmauer war; denn er hat mit 55-60 cm Höhe nicht nur die gleiche Höhe wie die nach links anschließenden Orthostaten, sondern die Fundlage zeigt, daß er im Zuge der eisenzeitlichen Bauaktivitäten lediglich nach vorne, über das Fundament in den Raum hinein aus seiner eigentlichen Position gehebelt wurde (Schachner 2014a, 106-107, Abb. 26‒27). Zwei vergleichbar bearbeitete sowie mit der gleichen Anzahl an Bohrlöchern und entsprechenden Abarbeitungen an den Unterseiten versehene Blöcke liegen seit den Ausgrabungen Th. Makridis 1911 auf der Schutthalde südlich von Kesikkaya.

Können wir mithin die Lage eines dieser Blöcke als Teil der Orthostatenmauer – etwa in deren Mitte – bestimmen, ist die Interpretation seiner bautechnischen Funktion in Ermangelung von Ansatzpunkten ‒ abgesehen von den Bohrlöchern ‒ schwierig. Möglicherweise deutet die große Zahl an Bohrlöchern daraufhin, daß nicht senkrecht stehende Balken auf dem Sockel standen, sondern daß kurze in Blockbauweise alternierend um jeweils 90° gedreht übereinandergesetzte Holzbalken einen beliebig hohen Pfeiler bildeten (Schachner 2014a: 107, Abb. 26).

Die Frage nach der Verwendung dieses Pfeilers läßt sich anhand der Befunde kaum lösen. Man kann spekulieren, ob dieser als Pfeiler oder eventuell auch als Sockel oder Unterbau für eine Statue oder ähnliches gedient haben könnte? Durch den Umbau des südwestlichen Teils des Gebäudes wird der Sockel unabhängig von seiner Nutzung obsolet, was auf einen funktionalen Wandel des Raums hindeuten könnte.

Mit den längsten, den Hethitern verfügbaren Baumstämmen hätte der Bereich 5/6 in seiner ursprünglichen Gestaltung, bei der auch der beschriebene Sockelblock seine Funktion erfüllte, kaum zur Gänze überdeckt werden können13. Es stellt sich mithin die Frage, wie dieses Areal in seiner ur-

<sup>13</sup> Die größte Breite des Raums beträgt etwa 9 m. R. Naumann gibt die maximale zu überdachende Spannweite hethitischer Räume mit maximal 8 m an (Naumann 1971: 134).

sprünglichen Gestalt zu rekonstruieren ist; zumal vergleichbare Orthostatenreihen nur an den sichtbaren Außenseiten von Gebäuden vorkommen14. Da die Orthostatenreihe als Verkleidung vor einem mindestens 4 m tiefen Sockel steht, dessen zweite Kante – falls überhaupt vorhanden – bisher nicht erfaßt wurde, ist es möglich, daß das Areals 5/6 nicht oder nur teilweise ‒ an den engen Stellen im Norden des Raums 5 (?) ‒ überdacht war.

Die Flucht der Mauer 10-719, von der bisher lediglich ein kurzes Stück freigelegt wurde (Abb. 10) (Schachner 2011a: 47, Abb. 27), zielt wie die der Orthostatenmauer auf die südwestliche Ecke des Korridors von Kesikkaya. Der zwischen 10-719 und 10-725 vorhandene Platz würde für eine ähnlich trichterförmige Gestaltung des Mauerendes wie im Falle von 10-725 / 10-719 ausreichen. Bei dieser Rekonstruktion ergäbe sich auf einer mindestens eineinhalb Meter höheren Stufe eine weitere Terrasse oder ein ähnlich dreieckiger Raum. Bemerkenswert ist, daß der Winkel zwischen den Mauern 10-748 und 10-725 ebenso 35-40° beträgt, wie der zwischen den Mauern 10-725 und 10-719. Es könnte sich also vor dem Westteil von Kesikkaya um einen in Terrassenstufen aufgefächerten Grundriß gehandelt haben.

Die Verkleidung des Mauersockels mit Orthostaten steht in einem deutlichen Gegensatz zu der bisher aus Ḫattusa bekannten Nutzung hethitischer Werksteine. Denn diese wurden stets als massive und das gesamte Gewicht der Mauer tragende Fundamentsockel eingesetzt15. Einzelne Orthostaten, die deutlich größer und zudem wohl bis zu etwa einem Jahrhundert jünger sind als die Beispiele bei Kesikkaya, sind an Toranlagen in Kuşaklı und in Ortaköy – hier bisher einmalig mit einem Relief – belegt16.

Im Gegensatz zur Verwendung einzelner Orthostaten an Toren ist eine vergleichbare Nutzung zur Verkleidung der äußeren Sockelzone eines Gebäudes in Zentralanatolien nur in Alişar Höyük nachgewiesen. Hier weisen die Sockel der südlichen und westlichen Außenmauern eines herrschaftlichen Gebäudes der alt-hethitischen Zeit (Schicht 10 T) senkrecht stehende, flache Steine auf (von der Osten 1937: 18, Fig. 10, 61, 65-71, Pl. 13, 17, 18). Im Gegensatz zu den Orthostaten an Kesikkaya sind hier jedoch deren Seiten und vor allem die Oberkanten nicht bearbeitet, so daß die Steine nur bedingt bündig stehen; Bohrlöcher zur Fixierung der Holzkonstruktion der Mauer, wie sie bei dem Gebäude vor Kesikkaya belegt sind, finden sich in Alişar Höyük nicht.

Die Nutzung sorgfältig bearbeiteter Steine als aufrechtstehende Orthostaten zur Verkleidung der Sockelzone von Mauern herausragender Gebäude ist im Gegensatz zu Zentralanatolien südlich des Taurus ein in der Mittelbronzezeit beliebtes Architekturmittel. Entsprechende, stets unverzierte Orthostatenreihen finden sich bei repräsentativen Bauten wie Toren, Tempeln oder Palästen unter anderem in den mittelbronzezeitlichen Schichten von Aleppo (Kohlmeyer 2000: Taf. 9), Ebla (Matthiae 1980: Fig. 24, 28), Tilmen Höyük (Duru 2003: 23, Abb. 12, Taf. 24.2, 25.1-2, 27.2, 28.1), Qatna oder Alalaḫ (Woolley 1955: 92, Taf. 13c, 15b).

Diese strukturellen und chronologischen Parallelen einer beim derzeitigem Forschungsstand in der hethitischen Hauptstadt wahrscheinlich an der Wende vom 17. zum 16. Jh. v. Chr. neu auftretende Form von Bauschmuck überraschen nicht vor dem Hintergrund der alt-hethitischen Expansion nach Nordsyrien (Klengel 1999: 33-84); vielmehr sie sind ein Beleg dafür, wie der indigene Grundriß eines repräsentativen hethitischen Gebäudes durch gestalterische Elemente der südlichen Kulturen ergänzt wurde. Bemerkenswert ist, daß weder Orthostaten noch andere architektonische Elemente aus Regionen südlich des Taurus bei Bauwerken der hethitischen Großreichszeit belegt sind17.

Die Funktion des Gebäudes ist aufgrund des Fehlens von *in situ* angetroffenen Funden nur hypothetisch eingrenzbar. Fragmente verschiedener Keilschrifttexte sprechen ebenso wie Teile großer Stierfiguren im weitesten Sinne für eine religiöse Funktion (Schachner 2014a: 109). Berücksichtigt man die Architektur wird deutlich, daß es sich aber nicht um einen Tempel gehandelt haben kann18. Im Hinblick auf den Grundriß des Gebäudes bestehen jedoch mögliche Parallelen ‒ die V-förmige Stellung der Mauern und deren gerade Abschluß ‒ zu den Kammern 1 und 2 in der Oberstadt (Abb. 13). Faßt man die Kriterien etwas weiter könnte man unter Umständen auch die häufig als NA4hekur oder Grab-

<sup>14</sup> S. unten S. 71.

<sup>15</sup> z. B. am Großen Tempel und vergleichbaren Kultbauten in Boğazköy oder beim Sphingen-Tor in Alaca Höyük.

<sup>16</sup> Kuşaklı (Südost-Tor): Schachner 1998: 130-132, Abb. 24-25; Kuşaklı (Nordwest-Tor): Mielke 2004: 148, Abb. 8; Ortaköy (Gebäude D mit Relief): Süel 2008: 34.

<sup>17</sup> Bereits die Torbauten in Kuşaklı und Ortaköy, die wenige Jahrzehnte jünger sind als das Gebäude südlich von Kesikkaya (s. oben Anm. 37), belegen eine deutliche Weiterentwicklung der Nutzung der Orthostaten. 18 allg. Neve 1999: 152-157, Abb. 72-73.

#### 72 *ANDREAS SCHACHNER*

stätte interpretierte Kammer B in Yazılıkaya in das Spektrum möglicher Vergleiche einordnen (Abb. 14)19 ‒ dies vor allem aufgrund der in beiden Fällen engen Verbindung zu einem Felsen.

Die enge funktionale Verbindung des Gebäudes zum Felsen von Kesikkaya ist aufgrund der Mauerfluchten, die auf den Korridor ausgerichtet sind, und der Anlehnung gegen den Felsen evident. Unklar ist beim derzeitigen Stand der Forschungen das Verhältnis zwischen dem Gebäude, dem Korridor und insbesondere einem Schacht, der auf der Ostseite des Korridors am Fuß der Felswand in den Boden eingetieft ist (Abb. 14) (Schachner 2014a: 110-112, Abb. 34-35).

Ohne hier die mögliche Parallelen des Korridors von Kesikkaya mit der Kammer B in Yazılıkaya und der Kammer 2 auf der Südburg im Detail diskutieren zu können, könnte man die Nennung der Errichtung einer «divine earth-road» (Hawkins 1995: 22-23, 44-45) am Ende der Inschrift in der Kammer 2 aufgrund der architektonischen Ähnlichkeiten zu dem Gebäude südlich des Felsen als möglichen Hinweis auf die Funktion des Korridors und des Schachtes sehen; trotz deutlicher Unterschiede erscheint eine Verbindung aufgrund der architektonischen Ähnlichkeit mit dem Gebäude südlich von Kesikkaya möglich, wobei eine formale Entwicklung dieses Gebäudetyps während der hethitischen Zeit und die Anpassung an die jeweilige örtlichen Gegebenheiten in Rechnung zu stellen sind20. In diesem Falle wäre zu fragen, ob das Gebäude südlich des Felsens im Verbund mit der einst vorhandenen Bebauung auf Kesikkaya und dem künstlich erweiterten Korridor mit dem Schacht in dieser Gesamtheit nicht einen Komplex für den Ahnenkult darstellen könnte? Im Falle dieser zugegebenermaßen teilweise spekulativen Interpretation würde der Gesamtkomplex von Kesikkaya und dem südlichen anschließenden Bauwerk vielleicht mehrere Gebäudetypen verbinden21, die in jüngeren Epochen in der Oberstadt von Hattusa räumlich getrennt voneinander errichtet wurden22.

Die Chronologie des Gebäudes ist ähnlich wie seine Funktion in Ermangelung gesicherter Funde schwer festzulegen. Die aufgrund der Anlehnung gegen den Felsen und die Ausrichtung der Mauerzüge offensichtlich enge Beziehung zu Kesikkaya, dessen westliche Hälfte wiederum mit der Poternenmauer verbunden war, legt den Schluß nahe, daß das Gebäude wahrscheinlich gleichzeitig mit der Befestigungsanlage an der Wende vom 17. zum 16. Jh. v. Chr. errichtet wurde23.

Eine, wenn auch aufgrund der Probe ‒ ein Holzkohlestück, das von zwischen den Orthostatenblöcken verbauten Balken herrührt ‒, nur vage Bestätigung für diese Interpretation ergibt sich aus einem 14C-Datum, das eine Errichtung des Bauwerks im 17. bis frühen 16. Jahrhundert andeutet (Tab. 2; Abb. 15). In Ermangelung weiterer Datierungen oder gesicherter Funde ist im Moment unklar, wie lange das Gebäude genutzt wurde.


Tab. 2 Radiokarbondatierungen aus dem Gebäude südlich von Kesikkaya

Die Errichtung dieses aufgrund seiner architektonischen Merkmale eindeutig öffentlichen, wahrscheinlich religiös24 genutzte Gebäudes etwa gleichzeitig mit der Poternenmauer hat Auswirkungen auf die Beurteilung der planerischen Gesamtentwicklung der Stadt. Basierend auf den Arbeiten in der westlichen Oberstadt gelang der Nachweis, daß in der zweiten Hälfte des 16. Jhs. v. Chr. durch die Verdopplung des Stadtgebiets eine wesentliche Veränderung des Gesamtcharakters der Stadt stattgefunden hatte (Schachner 2011a: 82-94). Der beschriebene Befund eines monumentalen öffentlichen Gebäudes südlich von Kesikkaya legt nun die Vermutung nahe, daß bereits deutlich früher und noch im topographischen Rahmen der sich seit dem späten 3. Jt. v. Chr. schrittweise, organisch entwickelten Altstadt

<sup>19</sup> Der dort vorhandene Bildschmuck spricht meiner Meinung nach nicht gegen diese Überlegung, da in Felsen gemeißelte Reliefs erst in der jüngeren Großreichszeit auftreten (Schachner 2012d). Der Korridor von Kesikkaya entstand jedoch mit großer Wahrscheinlichkeit bereits in (früh?) alt-hethitischer Zeit (s. unten). 20 Hawkins setzt die «divine earth-road» mit einem DINGIRKASKAL.KUR gleich (Hawkins 1995: 44-45).

Aufgrund dieses Texts wird klar, daß ein DINGIRKASKAL.KUR einen gebauten oder gestallten Raum darstellt. 21 Auf dem Felsen ein NA4hekur mit einem Gebäude für den Ahnenkult vergleichbar Kammer 2 auf der Südseite

von Kesikkaya. 22 In diesem Falle könnte die Bebauung auf Kesikkaya als NA4hekur angesehen werden, das später in der Oberstadt auf Yenicekale bzw. Nişantepe bestand (Schachner *im Druck* b; s.a. Schachner 2011c); das Gebäude südl. von Kesikkaya entspräche in diesem Falle in der jüngeren Großreichszeit den Kammern 1/2. 23 Für die bisherigen Argumente der Datierung des Gebäudes an die Wende vom 17. zum 16. Jh. v. Chr. s.

Schachner 2012a: 99-100.

<sup>24</sup> Siehe unten.

eine raumgreifende Umgestaltung nach übergeordneten, wahrscheinlich ideologischen/politischen Kriterien umgesetzt wurde. Diese umfaßte nicht nur funktionale Bauten wie die Befestigungsanlage oder den Getreidespeicher, sondern schloß den Naturraum (z.B. Kesikkaya, aber auch Büyükkaya, die Schlucht des Budaközü und Ambarlıkaya) in die Gestaltung ein und brachte indigene Monumentalarchitektur (z.B. das Gebäude südlich von Kesikkaya und die älteren Bauphase des Haus am Hang) hervor.

Betrachtet man die Gesamtentwicklung Hattusas von einer anatolischen Stadt zu einer nach repräsentativen Maßstäben gestalteten Metropole überregionaler Bedeutung zeichnet sich aufgrund der neuen Ergebnisse ab, daß der erste Schritt dieser Umgestaltung bereits unmittelbar nach der Etablierung der hethitischen Dynastie im Laufe des späten 17. Jh. v. Chr. erfolgte. Der Ausbau durch die Hinzufügung der Oberstadt im ausgehenden 16. Jh. v. Chr. stellt damit einen zweiten Schritt der Entwicklung von einer anatolischen Stadt zu einer repräsentativen Metropole dar.

Diese Entwicklung, die angesichts der beschriebenen archäologischen Ergebnisse wahrscheinlich bereits kurz nach Beginn der hethitischen Dynastie im 17. Jh. v. Chr. begann, ist eingebettet in eine Reihe technischer, wirtschaftlicher und sozialer Innovationen, deren Verbund diese Veränderungen erst ermöglicht (Schachner 2009). Gerade die Poternenmauer ist als Gesamtprojekt mit den ihr angeschlossenen Bauten ein Beispiel par excellence für die Verknüpfung dieser Innovationen: die Mauer an sich weist einen in dieser Zeit neuen Grundriß auf; die Errichtung der Mauer in einem äußerst anspruchsvollen Terrain belegt ein bis dahin ungekannt detailliertes Verständnis der Landschaft, in die das Bauwerk und die mit ihm verbundenen Gebäude auf verschiedene Art und Weise integriert werden (Seeher 2006c: 74-74, Abb. 39), und die technischen Fähigkeiten die Landschaft zu gestalten (Schachner 2011c); gleichzeitig wird die Landschaft auch selbst Teil des Bauwerks, wie das Beispiel von Kesikkaya eindrücklich zeigt; das große Getreidesilo auf dem Büyükkale-Nordwesthang steht für die technische Innovation der langfristigen staatlichen Kontrolle über große Mengen Getreide, die eine Säule des hethitischen Wirtschaftssystems darstellt (Schachner 2012b; Schachner 2009), während in dem monumental angelegten Gebäude südlich von Kesikkaya mit einem bis dato unbekannten Grundriß die soziale und ideologische Erneuerung zum Ausdruck kommt ‒ unabhängig von der Funktion, die das Bauwerk letztlich hatte.

#### *Zusammenfassung*

Die Forschungen der letzten Jahre in der Unterstadt von Hattusa tragen neben Antworten auf zahlreichen Detailfragen vor allem im Hinblick auf die generelle Entwicklung der Siedlung während der ersten Hälfte des 2. Jts. v.Chr. zu einem wesentlich verbesserten Verständnis bei. Unabhängig von ihrer Funktion wird deutlich, daß die Landmarke von Kesikkaya und deren Einbindung in das Stadtgefüge exemplarisch für die Kriterien und die Entwicklung der Gestaltung des urbanen Raums steht. Parallel zur Einbindung der Natur in den Stadtraum gelang der Nachweis indigener Formen monumentaler Architektur im Verbund mit technischen Innovationen. Diese Befunde zeigen, daß bereits im ausgehenden 17. Jh. v. Chr. mit einer, ideologischen Kriterien unterworfenen und großräumig angelegten Umgestaltung der Stadt begonnen wurde. Die textlichen Belege für diese Zeit sind nur schwach, können aber als Indiz dafür gewertet werden, daß diese bewußt gesetzten Veränderungen der bis dahin traditionellen städtebaulichen Entwicklung mit der sich in dieser Zeit neu etablierenden, hethitischen Dynastie und deren Institution zu verbinden ist (Abb. 16)25.

Bemerkenswert ist, daß bereits bei dieser frühen Umgestaltung der Stadt im späten 17. Jahrhundert v. Chr. grundsätzliche planerische Kriterien zur Geltung kommen, die später beim Ausbau der Oberstadt und anderen großangelegten Bauprojekten (z.B. die Gestaltung von Büyükkale [Schachner 2012c]) in weit umfangreicherem Maße sichtbar werden (Schachner 2011d; Dürr 2014; Schachner *im Druck* a). Merkmale wie die Einbindung und bewußte Gestaltung natürlicher Landschaftsmerkmale spielen ebenso wie die Errichtung völlig eigenständiger Architektur offenbar bereits vor der Errichtung der Oberstadt eine wesentliche Rolle bei der Planung der repräsentativen Gestaltung der Stadt.

Im Gegensatz zu dem durch staatliche Architektur geprägten Bereich südlich von Kesikkaya vermitteln die Grabungen nördlich der Landmarke einen Eindruck davon, wie sich die materielle Kultur in der ersten Hälfte des 2. Jts. v. Chr. fließend und ohne wesentliche Brüche über die politischen Veränderungen der Zeit hinweg entwickelt. Eine Umgestaltung wird erst durch die Errichtung eines Hauses

<sup>25</sup> KBo 3.57: Klinger 2006: 6-7.

(90) des frühestens im frühen 15. Jh. v. Chr. neu auftretenden Typs des Korridorhauses sichtbar. Diese Veränderungen beruhen im Gegensatz zu den aller Wahrscheinlichkeit nach durch den Staat "von oben" verordnete planerischen Konzeptionen auf der Südseite von Kesikkaya vermutlich auf sozialen Entwicklungen: die Einführung eines neuen Grundrißtyps spiegelt wahrscheinlich die Entstehung einer urbanen Oberschicht wieder (Schachner 2011a: 244‒250).

Erst durch die Errichtung des Großen Tempels mit dem angeschlossenen Südareal und der damit verbundenen Umstrukturierung der gesamten Unterstadt mit neuen Straßen und eventuell der Errichtung der Abschnittsmauer wird dieser Stadtbereich wahrscheinlich spätestens im Laufe der ersten Hälfte des 15. Jhs. v. Chr. den offiziellen Gestaltungskriterien vollständig und abschließend unterworfen. Zu dieser Zeit endet die Siedlungstätigkeit im Bereich nördlich von Kesikkaya, um den Blick auf die monumentale Architektur von der Stadtseite her ‒ d.h. von Süden und Osten ‒ freizugeben.

#### *Literaturverzeichnis*


Abb. 1 Überblick über die Unterstadt von Hattusa vor Beginn der Grabungen. Archiv der Boğazköy-Grabung (DAI, A. Schachner).

Abb. 2 Schematischer Plan des Areals nördlich von Kesikkaya (Stand 10/2013). Die hethitische Bebauung ist schwarz (jünger) und grau (älter) dargestellt, während die *kārum*-Zeit gelb markiert ist. Die roten Mauern bezeichnen die wahrscheinlich jünger-hethitische Zusetzung der Poterne 5. Die grau hinterlegten und gepunkteten Areale deuten die Flächen an, die – wahrscheinlich durch Suchschnitte der Grabungen bis 1912 – gestört waren. Archiv der Boğazköy-Grabung (DAI, erstellt von N. Strupler, nach Schachner 2014, 94, Abb. 2).

Abb. 3 Luftbild eines hethitischen Zentralraumhauses im Norden der südlichen Unterstadt. Archiv der Boğazköy-Grabung (DAI, A. Schachner).

Abb. 4 Die Bauphasen im Norden der südlichen Unterstadt. Archiv der Boğazköy-Grabung (DAI, A. Schachner).

Abb. 5 Baubefunde im Planquadrat 289/372 in der Sondage unter dem Fußboden von Haus 91. Archiv der Boğazköy-Grabung (DAI, A. Schachner).

Abb. 6 Graphische Zusammenstellung der kalibrierten 14C-Datierungen aus der nördlichen Unterstadt. Archiv der Boğazköy-Grabung (DAI, erstellt von N. Strupler).

Abb. 7 Graphische Rekonstruktion der Unterstadt in der hethitischen Großreichszeit. Archiv der Boğazköy-Grabung (DAI, P. Neve).

Abb. 8 Die Unterstadt von Hattusa von Süden. Archiv der Boğazköy-Grabung (DAI, A. Schachner).

Abb. 9 Luftbild des Felsens von Kesikkaya vor Beginn der Grabungen. Archiv der Boğazköy-Grabung (DAI, A. Schachner).

#### 78 *ANDREAS SCHACHNER*

Abb. 10 Die hethitische Bebauung südlich von Kesikkaya in der jüngeren Bauphase. Die dicken Linien markieren die erhaltenen Orthostaten. Archiv der Boğazköy-Grabung (DAI, S. Küçük).

Abb. 11 Der nördliche Teil der Orthostatenmauer. Archiv der Boğazköy-Grabung (DAI, A. Schachner).

Abb. 12 Luftbild der eisenzeitlichen und hethitischen Bebauung südlich von Kesikkaya; die Orthostatenmauer ist in ihrer gesamten Ausdehnung erkennbar; das hethitische Gebäude erstreckt sich in der unteren Bildhälfte; ober der Orthotatenmauer liegt rechts vor Kesikkaya die galatische Bebauung, an die sich nach links ein großes Gebäude der mittleren Eisenzeit anschließt; Norden ist rechts (Stand der Grabung 10/2013). Archiv der Boğazköy-Grabung (DAI, A. Schachner).

Abb. 13 Vergleich des Gebäudes an Kesikkaya mit der Kammer B von Yazılıkaya und der Kammer 2 in der Oberstadt. Archiv der Boğazköy-Grabung (DAI, A. Schachner).

Abb. 14 Gesamtansicht des derzeitigen Zustandes des Korridors an Kesikkaya von Norden; deutlich ist der Schacht vor der linken (östlichen) Felswand erkennbar (09/2013). Archiv der Boğazköy-Grabung (DAI, A. Schachner).

Abb. 15 Graphische Darstellung der kalibrierten 14C-Datierung aus dem Gebäude südlich von Kesikkaya. Archiv der Boğazköy-Grabung (DAI, A. Schachner).

Abb. 16 Schematische Darstellung der Stadtentwicklung von der *k*ā*rum*-Zeit bis in alt-hethitische Zeit. Archiv der Bogazköy-Grabung (DAI, A. Schachner)

# PLANNING A SACRED LANDSCAPE*.*  EXAMPLES FROM SARISSA AND ḪATTUSA

*Andreas Müller-Karpe*

#### *Abstract*

Mountain-Gods, sacred springs and other elements of landscape and nature in general, play an important role in Hittite religion. But in contrast to other early cultures, e.g. the Minoans, our knowledge about Hittite sanctuaries outside the cities is very limited. One rare example of such a sanctuary can be found up in the mountains south of Sarissa, Eastern Cappadocia. A special relationship existed between this sanctuary, the surrounding landscape and the town. This relationship is not only by chance but a planned one. The position and orientation of temples show a sophisticated plan integrating the landscape and also with an evident reference to the sun's path. The same principles can be seen in the capital and its surroundings.

A very typical feature of Hittite architecture is its special relation to the surrounding environment, the entire landscape. Rocky summits, steep slopes, crevices and other difficult areas were chosen as a building ground more often than ever before. But not only separate buildings show the attempt to integrate for instance rock formations; urban and regional planning also refer to landscape.

\*\*\*

Characteristics of Hittite planning can be seen very well at Kuşaklı-Sarissa, a site in the Upper Land, some 60km south of Sivas (Müller-Karpe 2002). The Hittite city is situated on a promontory at the southern edge of a high valley, called Altınyayla .The uppermost point of the acropolis has an altitude of 1665m above sea level, the surrounding mountains are about 2000m high. Correspondingly the climate is very rough, with short hot and dry summers and long winters with severe frost and heavy blizzards. Only two months in a year are complete free of frost. While the city flourished, environmental conditions had been a little bit better than today (Müller-Karpe 2009), but no doubt, is was an ecologically risky region.

In contrast to other urban centers, Sarissa did not emerge from a smaller settlement, its existence should be due to a royal foundation act. This act took place during Old Hittite Period in the early twenties or thirties of the 16th century BC.

But why just at this place? The region is not very fertile, no noteworthy resources and nowadays no important traffic routes. But geographically this landscape has a very special importance: The mountain range immediately south of Sarissa, the Kulmaçdağı, which should have been «Mount Sarissa» in Hittite Times, represents a crucial watershed: All watercourses north of this mountain range are running to the Kızılırmak (Marassanta/Halys) and finally into the Black Sea. But one half of the creeks and rivers south of the Kulmaçdağı runs to the Euphratus and finally into the Persian /Arabian Gulf, the other half is connected to the Seyhan and by that to the Mediterranean Sea (Fig. 1). This watershed divides Anatolia into three parts. It is the only place in the entire Near East, which has a hydrological connection with these three seas (Fig. 2). It is a geographical key position and by this fact it should have had also a symbolic, a sacral importance for the Hittite Empire. As we know, since the early times of the Hittite Kingdom it has always been the goal of expansion, to reach the sea, to «make the sea borders»1 .

<sup>1</sup> The Telipinu edict shows this quite clearly: Bryce 1998, 68 with further discussion of this expression. See also Puhvel 1957.

Another decisive element of Hittite geographical awareness are river systems. We know of the Seḫa-River-Land, the Ḫulaya-River-Land and also the importance of the Marassanta and so on (Arıkan 2007; Dinçol 1974; Garstang, Gurney 1959; Del Monte, Tischler 1978; Groddek 1997; Vieyra 1974). They defined even some political units.

But rivers originate in mountains and flow to the sea and in this context Mount Sarissa, as hydrological key point, should have played a special role, a role also in ideology of Hittite Kingship. And indeed, as it can be seen in CTH 591 (Fête du mois), Mt. Sarissa can be found exactly in a context within the principles of the ideology of Hittite Kingship, symbolized by the «Iron Throne» 2. Mt. Sarissa is mentioned at the second position on a list of mountains3 . In this text the function of these mountains can be interpreted so to say as basis of the throne, the support columns of the empire.

Even in times of the late empire Mt. Sarissa still maintained its importance. As David Hawkins (2006: 53) pointed out, Mt. Sarissa is mentioned in a context of *«*all the mountains of the Upper Land which My Sun [Tutḫaliya IV] habitually shoots*»*, i.e. he came here for hunting. Not really surprisingly for Hittite texts, the hunting of the Great King in mentioned in a religious text, an offering list, which enumerates also some river names.

Hunting should not be interpreted just as a holiday pleasure, but more a symbolic act, the subjugation of nature. This act at this mountain of all mountains might have had a special meaning, the subjugation of the vast territories connected by creeks and rivers to this mountain: Not less than nearly the entire Near East! Tutḫaliya IV could not find a better place to demonstrate his claim to be *šar kiššati*  'King of the whole (world)' (Klengel 1999: 294). He was the only Hittite King who used this title and the only one, whose visit at this mountain was mentioned. Furthermore in the district town Altınyayla/ Tonosa, just at the foot of the Karatonus Dağı, the highest point of the mountain range Kulmaçdağı, supposed to be Mount Sarissa, a sculptured stele with a relief was found, showing a person, probably the Great King, offering a libation to the God on a Stag, Kurunta (Fig. 3; Müller-Karpe 2003). There might be a direct connection to the erection of this stele with the visit of Tutḫaliya IV (Hawkins 2006: 63).

Mt. Sarissa as a religious and ideologically important landmark needed not only a stele but also a place of worship, a sanctuary up in the mountains. And this sanctuary needed a settlement at the food of the mountain for its subsist, its support and probably also for pilgrims. As to my opinion, this sanctuary is the Gölgediği or Suppitassu-sanctuary, 2.5km south of Kuşaklı up in the mountains (Müller-Karpe 1999: 79-86; Ökse 1999) and the related settlement is Kuşaklı itself (Fig. 4). So first of all, the landmark existed, the double watershed at the Kulmaçdağları. Second, the sanctuary Gölgediği was established, and third the city Sarissa was founded.

But why was the place of Gölgediği chosen to erect a sanctuary, and not the summit of the Karatonus Dağı? The answer might be, because Gölgediği has the only natural source-pond in the region (Fig. 5). Within a radius of at least 20km there is no other with a larger water expanse. It is a small karst lake, a former sink hole or doline, Turkish düden. Even today the water from its surface is flowing to the North, and finally into the Black Sea4 . But from its bottom it is possible, that some water may also flow to South – or at least the Hittites may have believed in it5 . DKASKAL.KUR, underground water courses, played an important role in Hittite religion (Gordon 1967). The source pond of Gölgediği/Suppitassu is the only place which has the condition precedent to be recognized as origin and connecting site of watercourses, leading to the three main seas surrounding the Hittite World.

As a part of the Kuşaklı-Sarissa project, Tuba Ökse (1999: 88) excavated the effluent of the pond. She was able to proof, that this effluent had been framed artificially by slab stones. Even without any excavations a pavement at the bank of the small lake can still be seen at the surface. In Hittite Times the water level had been raised artificially. To prevent the loss of water through percolation, a circular ditch was dug in some meters distance to the bank (Fig. 6). Then the ditch was filled up with a special impermeable, so called aquifuge, clay. The same technique was used in context with the erection

<sup>2</sup> «Auf, Berg Sarissa, erhebe dich! […] Die gute Nachricht soll sie finden, unsere mächtige Sonne [Majestät] (und) die Tawananna auf einem Thron aus Eisen» (Klinger 1996: 321, 362). 3 1. Mt. Puskurunuwa, 2. Mt. Sarissa, 3. Mt. Tutḫalija, 4. Mt. Arnuwanda (Klinger 1996: 320-323).

<sup>4</sup> Nowadays water flows only temporarily and dries up during summer. Its natural course leads first to Sarissa, then via the Karaboğaz Deresi north of Altınyayla/Tonos and the Üsgülüç Deresi, passing Hanlı to the Kızılırmak (Maraššanta/Halys). It is worth mentioning that the mouth of the creek 'Üsgülüç Deresi' on the Kızılırmak is in the vicinity of Kayalıpınar/Samuha. This means there was a direct connection between the two Hittite cities. Probably Sarissa was founded from Samuḫa.

<sup>5</sup> Such an idea is not as absurd as is seems to be at first sight: Even the Rhine and the Danube have some common sources.

of water dams, as we learned from the excavations down at Sarissa (Hüser 2007: 80, 81, 111, 124) and Boğazköy (Seeher 2006: 19-21).

In the surroundings of the pond a lot of traces from buildings and installations were found. The most important ruin is a temple on top of a promontory to the west above the pond (Fig. 6). It is not excavated yet, but the main elements of its structure had been investigated by a geophysical survey. The building is orientated exactly North-South. Like other Hittite temples, its central element was a courtyard, framed by rooms. Only a staircase, leading to a lower terrace, shows a little difference in orientation. But in a downward northern extension of the longitudinal axis of this temple, the city Sarissa is located in a distance of 2.5 km and 300 m lower at the foot of the mountain range. As to my opinion, the connection between the two sites is quite obvious. The temple up in the mountain and the city down at the edge of the valley were planned together. They complement one another. It is more than only by chance, that the orientation of the temple and the visual axis, leading to the city, are on the same line. This feature indicates a discreet planning of the 'sacred landscape' according to the cardinal points of the compass.

These principles of planning can also be found at the city itself. Building C at the South-Eastern slope of the acropolis, presumably the temple of the storm god of Sarissa, is also oriented according to the cardinal points of the compass (Fig. 7). In this case the axis of the building is not aligned North-South, but the corners of the building point to the extremities North, South, West and East. This means, the axis of the building is rotated in relation to the North-South line. The mid between the cardinal points had been chosen for the building axis. The rotation represents not exactly 45°, but 44.3°, as we learned from precise measurement. Thus only a minor deviation has to be noticed.

Also other public buildings in Sarissa show the same orientation like building C. There are good reasons to suppose, that this temple had been the first one to be built, when the city was founded and correspondingly it had been the model of orientation for other buildings. Only public buildings show this specific orientation. It was not obligatory for domestic houses.

The position of the city gates however is undoubtedly also connected with these planning axes of the big temple. It is obvious, that the basic structure underlying the plan of the city were these diagonals to the cardinal points of the compass (Fig. 7, light blue, dotted lines). The city was founded and erected according to this geometric, orthogonal master plan. Later buildings deviate from its orientation.

The planning axes of the Old Hittite outline of Sarissa can also been found outside the city: Dams for water reservoirs had been erected according to the master plan. Even in greater distances to the city some pathways still in use and margins of actual fields show the same orientation (Fig. 8). It seems, that traces of an Old Hittite allotment or parceling has survived in this region (Müller-Karpe 2013: 343). This means, that we have here an example for a planned and artificially structured landscape. The allotment should be dated to the Hittite Era, because of its congruence with the master plan of Sarissa and the fact, that this region did not play any major role after this era. We can call it a partially fossilized landscape with substantial Bronze Age remains. The allotment in the high valley «Altınyayla ovası», the surroundings of Sarissa, anticipates orthogonal field systems in Greek and Roman colonies (*centuriatio* or *limitatio*).

But even within the city *Sarissa*, not all Hittite structures follow this orientation pattern. Some buildings from the old Hittite time show an obvious different orientation. The most prominent example is the temple on the north terrace. According to the dendro-chronological analysis of timbers, this building already belongs to the foundation phase of the city (*c.* 1530 BC). Is there a special significance of this striking difference in the orientation? As a sacral construction presumably nothing was left to chance. The orientation should have been chosen with consideration. This specific orientation fits astonishingly well to the summer solstice in the late 16th century BC, with only a slight discrepancy of less than one degree. We can say, the longitudinal axis of the building facing to North East had been aligned to the most northern rising point at the horizon (Müller-Karpe, Schrimpf 2009). The sun reaches this point around June 21.

It is also possible, that the Hittite architect chose the most northern turning point of Venus as morning star, but the planet reaches this position only every 8 years. We have to take a connection with the course of Venus into consideration, because this building probably was an Ishtar-temple. Looking along the longitudinal axis of this building to the opposite direction, facing South West, the angle of the outline fits well to the setting point of the sun at the day of the winter solstice. Pure coincidence is extremely improbable.

The north eastern city gate and the temple at the northern terrace were planned together, they have exactly the same orientation. Beside this temple also other, still unexcavated buildings detected by geomagnetic surveys, show this specific orientation leading to the rising point of the sun on summer solstice. Correspondingly the northwestern city gate points to the setting point of the sun at the same day. So the northern district of Sarissa seems to be planned according to the path of sun, especially according to the summer solstice (Fig. 7, yellow dotted line).

To summarize the main features of planning at Sarissa.


It becomes apparent, that the Hittite planners tried to integrate the main features of the cosmic order into the outline of Sarissa and its surrounding landscape. The outline reflects the path of the sun.

This cosmic order can also been drawn as the wheel of the solar year (Fig. 9). The year is divided into two halves by the two solstices and also by the equinoxes. Correspondingly every year has four seasons. The time between a solstice and an equinox is always around 90 days, at the wheel of the year 90 degrees, a right angle. The exact numerical value differs a little bit, as we know today. But this was presumably not important for the Hittites.

Every right angle can be divided into two halves, the diagonals equal to 45 degrees or 45 days at the wheel of the year. And these diagonals are the most important axes for the planning of Sarissa and the land divisions in its vicinity, the entire Altınyayla ovası. This grid gave the landscape its artificial structure, which ignores often the direction of slopes. The specific orientation of this grid had no practical relevance; it should have had a sacral meaning.

In Ḫattusa the situation is more complicated, because this city developed gradually during several centuries from a smaller settlement to one of the biggest capital cities of the ancient world. We cannot expect a master plan for the entire city, comparable to that one of a new founded settlement like Sarissa. But the largest district, the so called Oberstadt (Upper City) shows a distinctive, well thought-out plan. The strange geometrical outline of the Upper City had certainly already struck Karl Humann in 1882. Peter Neve (1992: 22) emphasized that there is an axial connection (North-South axis) between Yerkapı und Nişantaş. He also reconstructed a 'via sacra' passing the main city gates and showed, that the city wall, *Yerkap*ı and the gates had not only a function as fortifications but also a representative and religious meaning.

But how was the Upper City planned and erected? What is actually the basic pattern, the leading idea behind the outline of this district? The leading idea behind the outline is simple, as well as brilliant. Supposedly we can reconstruct the planning of the Upper City of Ḫattusa as the following: The starting point should have been the highest point of the terrain. Now it is the Sphinx Gate, but before the erection of Yerkapı, most probably the spot in front of the northern entry to the tunnel had been the highest place. From here the Hittite city planner observed – let's say about 1500 BC – at the day of the summer solstice early in the morning at the horizon the rise of the sun (Fig. 10). On this visual axis an assistant put some pegs to mark the place, where the King's Gate should be constructed. In the same way on this axis the place was pegged out, where temple 5 was planned to be erected. This building shows exactly the same orientation. A coincidence is very implausible. At the evening of the same day, a 21 June, the Hittite city planner observed from the same place at Yerkapı also the sunset and his assistant pegged out the place, where the Lions Gate should be erected.

The city planner returned to the starting point, exactly six months later, at the day of the winter solstice, 21 December. He came here, in order to mark the building site for the big bastion with the Sphinx Gate. Again with the aid of at least one assistant, he marked the axis of the sunrise and in the evening the sunset. Right on these lines are the basic corners of the mighty building (Fig. 11).

Its longitudinal axis follows also very precise the East-West line, which corresponds to the rising and setting points of the sun at equinox, around 21 March or September. The lines toward the rising and setting points of the sun at the winter solstice had only been important for planning the building, not for any further observations after finishing it. These axes have just a symbolical meaning, not a practical one. The path of the sun is, so to say, 'inscribed' into the building (Müller-Karpe 2013: 345-351).

Another angle is also worth mentioning. The lines to the rising and setting points form together with the East-West line, or equinox line an angle of 32 degrees. It is quite astonishing that we can find the same angle of 32 degrees also in the vertical view. It is the gradient angle of the bastions stone pavement at Yerkapı! I have no explanation, as to what this means, but I am very much convinced that it means something.

This demarcation of the central 'cult and temple city' of the Empire was more than just a city wall. In the planning of this wall a connection with the path of the sun is evident, but it is not just a calendar building or observatory. The main idea of the planners was to monumentalize the cosmic order, and to build the city in accordance with the divine structure of the world. Ḫattusa as the 'City of the gods' has been planned to fulfill its function as central element of a landscape, seen from its inhabitants as a sacred one.

The planning of the sacred landscapes of Sarissa and Ḫattusa are also two examples for the effort of Hittite land surveyors, most probably priests and architects to translate time (the circle of the year) into space.

#### *Bibliography*


Hüser A. 2007. *Hethitische Anlagen zur Wasserversorgung und Entsorgung.* (Kuşaklı-Sarissa 3). Rahden.


Fig. 1. Watercourses in the vicinity of Kuşaklı-Sarissa and the watershed between the Black Sea, Mediterranean and Persian/Arabian Gulf (marked in red).

Fig. 2. The Hydrological key point of Anatolia nearby Kuşaklı-Sarissa.

Fig. 3. The stele from Altınyayla at the foot of Mt. Sarissa.

Fig. 4. Sarissa and the Gölgediği/Suppitassu-Sanctuary.

Fig. 5. Suppitassu-Sanctuary at Gölgediği, south of Sarissa, with its source-pond.

Fig. 6. Plan of the Suppitassu-Sanctuary at Gölgediği, south of Sarissa.

Fig. 7. Sarissa in Hittite Times. Dotted lines: Planning axes from the foundation phase (late 16th century BC).

Fig. 8. Sarissa and its surrounding landscape, with traces of an Hittite allotment.

Fig. 9. Reconstruction of the wheel of the Hittite year. Fig. 10. The Upper City of Ḫattusa, with a reconstruction of the planning axes according to the path of the sun.

Fig. 11. The southern bastion with the sphinx gate in Ḫattusa and a reconstruction of the planning axes.

# THE SACRED LANDSCAPE OF SARISSA

*Gernot Wilhelm* 

#### *Abstract*

Some fragments from Sarissa have become particularly important for the reconstruction of the sacred landscape of the city (Kuşaklı, province of Sivaş). They contain the description of a spring festival celebrated by the king. There are also fragments from Ḫattusa that concern the celebration of festival in this peripheral town. A comparison between texts and the sacral landscape of Sarissa, as it was reconstructed during excavations, offers a rare chance of identifying cultic installations and itineraries. Further excavations, especially by the Sarissa *ḫuwasi* complex would be a particularly important project.

\*\*\*

Among about sixty cuneiform tablets and tablet-fragments excavated at various locations within the walls of Sarissa (Kuşaklı, province of Sivaş), fragment KuT 19 has become particularly important for the reconstruction of the sacred landscape of the city. It was found in 1994 in the so-called 'tablet room' of Building A (Müller Karpe 1995: 25-27; 1996: 74-83) on the western rim of the city and it was published the following year (Wilhelm 1995: 37-38; Wilhelm 1997: 17-18, Tafel 1 and 21).

Fragment KuT 19 is the upper left corner of a tablet written in the late Empire period. It contains the first lines of the agenda of the spring festival, which was celebrated by the King when he came to Sarissa. Only 15 lines are fairly well preserved, but fortunately they yield important topographical information.

The first two paragraphs read: «When in springtime the king goes to Sarissa in order to celebrate the festivals –, as soon as the king approaches the city, he does not go up to the city, but the king follows the upper road up to the *ḫuwasi* of the Storm-God».

The next paragraph mentions the «lord of Sarissa» and again refers to the *ḫuwasi*.

Another fragment, which resembles the first in size and form of script, describes the cultic activities early in the morning of the 3rd day of the festival. Most of the preserved text contains standard phrases, the main information again is the reference to cultic activities at the *ḫuwasi*:

When on the 3rd day it is getting light, they open the *ḫalentuwa*-building at Sarissa. They draw open the curtain. The king enters the bath-house and takes the robes and the ceremonial vestments. The king leaves the bath-house. Two palace attendants and one body-guard (LÚ *MEŠEDI*) walk in front of the king. The king steps on the chariot and drives up to the *ḫuwasi* of the Storm God. In the *ḫuwasi* sanctuary of the Storm God […].

Only very little of the 4th column on the reverse of the tablet is preserved, but it is enough to conclude that the tablet closed with the last cultic acts of that day and that a fourth tablet was to follow.

The festival at Sarissa had been known even before the discovery of these texts. Emmanuel Laroche (1971) has an entry «Fête à Saressa(?)» in his catalogue (CTH 636) under which he subsumes several large fragments of the ritual for a festival with reference to Sarissa. One of them presumably gives the agenda of the second day, another one, which is the last of the series, contains the agenda of the 4th day (Wilhelm 1997: 10-14). Oddly enough, we have parts of the 1st and the 3rd tablet from Sarissa itself, and the 2nd and 4th from Ḫattusa.

The tablets from Ḫattusa give additional information on the topography of Sarissa. The 2nd tablet (KUB 20,99) again describes the king's cultic performances at the *ḫuwasi*, to which he drives up on his chariot. We learn that there is a *ḫilammar* «gate building» or «porticus» and that the king enters and

#### 94 *GERNOT WILHELM*

approaches the *ḫuwasi*. In the vicinity of the *ḫuwasi* of the Storm God there is a *ḫuwasi* of the Stag-God, d KAL, who receives offerings together with his consort Ala. Close to these *ḫuwasi*s, there is a sacred spring called Suppitassu, which also receives offerings.

Having finished his ritual duties the king apparently returns to the city and stays there overnight. In the morning of the following day he then drives up again. On the 3rd day, as we have seen, the king first visits the building or enclosure that houses the *ḫuwasi* of the Storm God. On the 4th day after leaving town he drives up in order to visit the *ḫuwasi* of the goddess Anzili1 .

The ritual of the 4th day is important because it contains a reference to a *ḫalentuwa-*building next to the *ḫuwasi*-compound2 . In the past, it was controversial what kind of a building a *ḫalentuwa* actually is*.* H.G. Güterbock (1974: 307-311, 313) defined it as a palace, whereas S. Alp (1983) tried to prove that it was the «temple» or «adyton».

The Sarissa ritual refers to two different *ḫalentuwa*-buildings, one in the city and another one in connection with the *ḫuwasi-*complex. This situation suggests that a *ḫalentuwa* is neither a palace nor the adyton of a temple, but a location where the king may spend the night and where he is able to perform certain cultic acts and to put on the attire required for celebrating religious festivals.

When staying in Sarissa overnight, the king sleeps in the *ḫalentuwa* in the city. The lord of Sarissa mentioned in Tablet I of the ritual certainly resided in an appropriate building within the walls of his city, and a good candidate seems to be Building E on top the acropolis (Müller-Karpe A. 2002: 336-7; Arnhold 2009). This building was erected at the end of the 16th century BC and destroyed approximately a century later. After a period of decay it was rebuilt in larger dimensions (42x12 m). Though its function could not be determined, ceramic finds point at a non-cultic use (Arnhold 2009: 125 and 135).

It is, however, well conceivable that the special religious status of the king, his carefully preserved purity, made it impossible for him to spend the night in the residence of his subordinate. It is possible then that the king stayed in a special suite of a temple compound, being, so to speak, the guest of the god for whom and with whom he celebrated the festival. Tentatively, one might think of the temple on the northern terrace at Kuşaklı (Fig. 1) (Müller-Karpe A. 1995: 9-21). This temple was built in the second half of the 16th century BC and suffered heavy damage like the rest of the city in the early 14th century. Unlike the larger Temple C, however, it was repaired and still in use in the empire period, though, according to the ceramic analysis (Müller-Karpe V. 2006), not as late as the late 13th century. The tablets of the Sarissa ritual from Boğazköy were written in the time of Tutḫaliya IV, but the composition of the ritual might have taken place earlier. The large Room 51, which is part of the enclosure of the temple-courtyard, is connected with an annex that seems to have been a bathroom (Rooms 27/28). Both a room for staying overnight and a bathroom are prerequisite for the king's preparation for ritual acts according to the standard form of festival rituals. It seems well possible that these rooms were used by the king during the festival period. In this case, they would have functioned as a *ḫalentuwa*.

Evidently, the *ḫalentuwa-*building in the *ḫuwasi-*compound is not used for spending the night there. Its location in relation to the *ḫuwasi*s is illustrated by the itinerary of the king on the 4th day of the festival: he drives up by chariot to the *ḫuwasi* of the goddess Anzili. In the gate-building (*ḫilammar*) he gets off the chariot and enters the *ḫalentuwa-*. There he breaks a loaf of bread and performs a libation, he then leaves the *ḫalentuwa-* and approaches the *ḫuwasi* of Anzili. Hence, the *ḫalentuwa-* should be close to the entrance of the *ḫuwasi* complex. The following line tells us that the *ḫuwasi* of Anzili stands inside a building or an enclosure and is only accessible by passing a gate (ká).

The complex of a circular pond (Fig. 2), remains of buildings and other peculiar features discovered in the mountain range 2.5 km south of Kuşaklı in 1996 (Müller-Karpe 1997: 118-120; 1999: 79- 86), suggests an obvious parallel between topography and texts. The circular pond reminds one of the sacred pond Suppitassu mentioned in the Sarissa ritual.3 A geophysical survey and a small excavation suggested that the pond was surrounded by a wall made of mud-bricks (Stümpel *apud* Müller-Karpe 1998: 152-153) and that it had an outlet artificially framed by stones (Ökse *apud* Müller-Karpe 1999: 86-91). The rectangular structure on the east side of the pond (Fig. 3) has been interpreted as a podium (Müller-Karpe 1999: 85 sub G) from which the sacral pond may have received the offerings mentioned in the Sarissa ritual. Taking into consideration that both the pond Suppitassu and the Stag-God's *ḫuwasi* jointly receive offerings, it is likely that the *ḫuwasi-*building of the Stag-God was not far away.

<sup>1</sup> For the identification of Anzili and d*IŠTAR(-li)* see Wilhelm 2002: 348; 2010b.

<sup>2</sup> KUB 7.25 i 7-14, ed. Wilhelm 1997: 13.

<sup>3</sup> Wilhelm 1997: 11-12, KUB 20.99 ii 14; 22.

According to two letters (Wilhelm 1998; Mouton 2007: 20; Hoffner 2009: 262-267; Wilhelm 2010a: 378-379) of the early 14th century found in Temple C, Sarissa seems to have enjoyed the fame of being a prominent place for obtaining bird oracles. The pond is a place favoured by birds, as this author was able to observe, and it seems quite possible that here, in the presence of gods, was the right place where birds could be watched for divination.

West of the pond, on a hill which was artificially flattened to form a terrace, remains of a rectangular building of considerable dimensions (75x45 m) were detected (Fig. 4). The space within the walls is at least partially paved, and so is the slope. The access was in the north. As there are no traces of another building of similar dimensions, it is likely that it is the *ḫuwasi-*sanctuary of the Storm God.

In a prominent central position within the space between the walls of the longer sides there is a larger block of stone, about 2 m high, and a smaller one next to it (Fig. 5). Both blocks are unworked, and thus there is no chance of dating them or at least of proving that they had already been there when the temple was still in use.

May we assume that these blocks were the *ḫuwasi-*stones mentioned in the ritual? Quite some research has already been done on the shape and function of a *ḫuwasi* (Darga 1969; Popko 1978: 123- 127; Güterbock 1983: 215-217; Hutter 1993; Haas 1994: 507-509). The main results are the following:


Despite the abundance of attestations in texts, no *ḫuwasi* has ever been identified with certainty4 . We cannot be sure that the big blocks in the centre of the temple are indeed *ḫuwasi*s. They seem to have broken off from the steep rock-face and then may well have been regarded as a visual proof of the powerful raging of the Storm God; but this remains speculation.

It seems that there were more *ḫuwasi*s in the sacred landscape of Sarissa. Fragment KuT 54, excavated in 2001, is a list of offerings for the *ḫuwasi*s of at least ten gods. Unfortunately the names of the gods are all broken off.

On a lower terrace to the north of the *ḫuwasi* sanctuary of the Storm-God remains of another sizeable building are visible (Fig. 4). According to the geo-electric survey, undertaken in 1998, it is subdivided into several rooms (Lorra and Stümpel *apud* Müller-Karpe 1999: 108 [E\_1]). Most likely this building is the *ḫalentuwa*-house because according to the texts this is the first building that the king visits after having reached the *ḫilammar* and before approaching the *ḫuwasi*s.

The modern visitor usually approaches the sacred area from northeast. The «Upper Road» to the *ḫuwasi*s, however, which the king takes before entering the city, presumably branched off the mainroad somewhere west of Sarissa. This would better fit the western location of the *ḫalentuwa*-house. In 1997, the palaeobotanist of the excavation, Rainer Pasternak, and the present author tried to identify the road by leaving the *ḫuwasi*-sanctuary at its western edge, using Mr. Pasternak's car. We indeed managed to get to the main road without too much difficulty (Fig. 6). There was only one bad part, but the track could have been repaired without much effort.

The sacred landscape of Sarissa in connection with the ritual of the spring festival offers a rare chance of identifying cultic installations and itineraries. Excavating the Sarissa *ḫuwasi* complex would be a particularly important project.

#### *Bibliography*

Alp S. 1983, *Beiträge zur Erforschung des hethitischen Tempels. Kultanlagen im Lichte der Keilschrifttexte* (Türk Tarih Kurumu Yayınları 6/23), Ankara.

Arnhold S. 2009, *The Hittite building E on the Acropolis of Kuşaklı* (Kuşaklı – Sarissa 4), Rahden/Westf.

<sup>4</sup> For the suggestion that the rock sanctuary Yazılıkaya was a *ḫuwasi*, see Singer 1986.


Popko M. 1978, *Kultobjekte in der hethitischen Religion (nach keilschriftlichen Quellen)*, Warszawa.


Figure 1. The temple on the northern terrace at KuŞaklı (after Müller-Karpe, A., 1995, 10).

Figure 2. The circular pond 2.5 km south of Kuşaklı seen from the south-east (author's photograph).

Figure 3. Map of the pond area. (after Müller-Karpe, A., 1999, 81).

Figure 4. The upper plateau with remains of a temple seen from the south (author's photograph).

Figure 5. The blocks of rock on the upper plateau seen from the north-west (author's photograph).

Figure 6. The alleged course of the 'Upper Road' branching off the main road in the west of Kuşaklı and leading upwards to the *ḫuwasi*-sanctuary (after satellite photograph of Google Earth, 26th January, 2014).

# THE RELIGIOUS SIGNIFICANCE AND SACREDNESS OF THE HITTITE CAPITAL CITY SAPINUWA

*Aygül Süel*

#### *Abstract*

The Hittite state governed by a theocratic monarchy in 2nd millennium BC deeply influenced its age. Documents related to birth, death, purification, throne accession, building, communication with the gods, epidemics in army, etc. have a significant relevance in the Archives of the Hittite Kingdom.

Archaeological evidence from the Hittite city Sapinuwa is a proof about the performance of these rituals during the Hittite era, and a great contribution to the study of the sacral role of the city itself. In this respect, Sapinuwa is not only a capital of the Hittite State but also very significant religious centre of the Hittite World.

The great kings ruled the Hittite kingdom in name of the gods. The thousand gods mentioned in Hittites texts are usually understood as a sign of the great tolerance of the Hittites. Most of all, I think, that the Hittite king purposed to obtain the power of the gods of the subjected populations by getting them into his state pantheon. In this way, he could rule over the subjected kingdoms in the name of these gods.

\*\*\*

Documents related to birth, death, purification, throne accession, construction of new buildings, communication with the gods, cleansing of mouth, domestic violence, epidemics in army, etc. take great importance in the Hittite State archives. The ones found in Ortaköy show the strong sacred function of the city during the kingdom of the King Tutḫaliya III and his wife Taduḫepa.

#### *1. The Administrative Importance of Sapinuwa*

The Hittite city Sapinuwa is located in Ortaköy, 55 km southeast of modern Çorum. This ancient city was one of the capitals of Hittite Kingdom and, having almost 9 km2 of expansion, was one of the large scaled towns in 2ndMillennium BC.

Expressions in official letters, such as: «Tell my master, the Great king! Everything is all right in the town; it is defended from the enemies», point out that the administrative territory of Sapinuwa included many towns and covered a huge region.

According to the Ortaköy texts, the royal couple, Tutḫaliya III, the Great King, and his wife, Taduḫepa, ruled in Sapinuwa in 14th century BC. After the death of Tutḫaliya III, his son, Suppiluliuma I, also ruled the Hittite Kingdom from there.

The city is not only a capital of the Hittite State but also a very significant religious center of the Hittite world.

#### *2. The Religious Importance of Sapinuwa*

Hittites believed that gods controlled their destinies, and that they endow all good and bad things in their lives, such as illnesses, defeats, etc. In order to satisfy them, they offered various kinds of animal sacrifice, and they strain to purify themselves by prayers and festival rituals. They also made effort to understand the requests of gods and the reasons of their anger through fortune telling.

#### 102 *AYGÜL SÜEL*

The rituals among the Hittite text corpus, performed in order to purify and to cleanse the potential pollution, represented an ancient tradition. One of the most important of these rituals is *itkalzi*-ritual (Fig. 1).

*Itkalzi*-rituals were especially performed to purify the royal family. They included ceremonies, named by purifying agents. The order of the ceremonies followed like that: the words of water, the words of oil, the words of silver, the words of lapis-lazuli etc.

An example of purification ceremony performed with water could be the following: «And he (the one who offered the sacrifice) washes himself. Yet when he prepares himself to wash, the priest, who keeps the pure water, bring them washing […] and when the one, who offers the sacrifice, finish to wash himself, they pure this water into a copper or bronze vessel».

It was suggested that *itkalzi*-purification rituals were performed in Building D of Ortaköy. The purified person (Hittite king or Hittite queen) probably washed himself in the basin (it was ruined, but the system of wastewater discharge remains still today) in the middle of this building (Fig. 2). This ceremony must be symbolic. Behind the place of the basin was a wooden wall now revealed only by traces on the burned plaster.

On one orthostat, posed in the entrance of Building D was depicted the God Teššub. The god with armors, resting against a spear, welcomes with his left hand the ones who enter the building (Fig. 3).

Bronze spear heads and axes which were inscribed with the title «The Great King», a bronze armor, and a bronze helmet, and the impressions of royal seals, discovered in the corner of a room in this building, provided us extremely significant and interesting information related to these rituals (Figs. 4-5).

One of the cuneiform tablets from the Ortaköy archive belongs to an *itkalzi*-ritual written on the occasion of the marriage of Taduḫepa and Tašmišarri; a purification ceremony, performed by Taduḫepa, was performed in this context.

Several cuneiform tablets about these ceremonies are known and were unearthed in Sapinuwa. As far as we know from cuneiform sources, a ritual ceremony was performed in Zithara. The original tablets were imported from Sapinuwa. Expressions like «We reproduced these tablets from the originals in Sapinuwa» which are mentioned in the colophons of Boğazköy texts, suggest that Sapinuwa was an important religious center for the Hittites.

A further aspect of these tablets from Sapinuwa is that the clay was extremely well-prepared, and the script was elegant and correct. These cuneiform tablets were probably written with a golden stylus, a rare find for Ortaköy excavations (Fig. 6).

One of the Hittite-Hurrian texts from the city mentions the pure water supplies for the rituals: three of from 7 vessels of water were supplied from the main water spring of Sapinuwa, one from main water spring of Šulupašši, one from the spring in the vineyard of Sapinuwa, one from GN-river and one from the river Šapantalliya (Fig. 7).

During Ortaköy excavations, many religious and administrative buildings have been discovered (Fig. 8).

Building A, where royal seals were discovered, is a monumental structure with a unique symmetrical architecture. Several cuneiform tablets from the Hittite period were unearthed in this monumental building.

150m east of this structure, the remains of the Building B were unearthed. Remains consist of sewing tracks on a cyclopean foundation, and mud-bricks wall of 1.5 m in height. This building covering a 1250 m2 has the appearance of a depot. Almost 70 jars, which have nearly 1-1.5 tons of storage capacity, were discovered there.

#### *3. Religious Structures*

The remains of buildings near the southern terrace were identified as Building C and D. Spear heads and axes with a cuneiform inscription «Great King», unearthed together by the gate of Building C, suggest that it was probably a royal complex. Finally Building D is a very important religious structure for Sapinuwa.

Remains of fortification walls (almost 75 m length of it), surrounding the area including these monumental structures, were unearthed.

Another area with religious importance was discovered just outside in Ortaköy is the Tepelerarası region. This broad area is called «Ağılönü».

Ağılönü covers an area of thirty thousand square meters and recalls a rectangular peninsula lying from north to south; it is surrounded with deep gullies from the west, the south and the east.

The Ağılönü excavations began in the eastern side of this region in 2000. During the excavation, a street paved with stones lying in north-south direction and a few workshops were discovered. In the entrances of these workshops alongside the street, millstones and ovens were unearthed. A *pisé* wall, built with clean-cut mud-bricks of 40x40cm, was located between the street and the side of the hill overlapping the side of Ağılönü hill.

A number of terracotta artifacts and utensils were unearthed in this complex where Building 1 and 2 were excavated (Figs. 9-10). In the west side of Ağılönü was excavated a trench in which a huge architectural structure, probably an over, was unhearted (Fig. 11). In southwest of this oven, a lot of bones and terracotta artifacts were discovered into a cesspit. Undamaged offering vessels, fragmentary utensils and a naked goddess figure (Fig. 12) were also discovered in this archeological area.

The most important artifact in the Ağılönü region is a stone pavement covering almost 2000 m2 (Fig. 13). In the northern side of this stone pavement there was a canal filled with rubble, on the other three sides were built the retaining stone walls. This structure survived in twelve layers, despite thousand years of damage, until today. Despite the heavy wash causing by agricultural machines the general structure of this stone pavements was preserved. Just some surface stones beneath the upper surface of field were pulled out by ploughs, especially where the ploughs were pulled in the east-west direction over the area. For this reason the surface has a scalar appearance.

When we concentrated to the southern side of the stone pavement the comparatively intact part of this structure showed that this stone pavement was built in form of a rectangular prism, not in a scalar form, probably in order to keep away the rain and the snow. This area was isolated from the center of town by a deep gully. The massive walls surrounding the stone pavement brake off this area form the outer world.

A number of offering vessels, vases with base, jugs and *rytha* were unearthed over the ground level of the trench. Among the artifacts, two lion heads are especially distinctive. Produced in different styles, both decorated the mouths of long necked jugs.

They were probably used for the performance of ritual ceremonies (Figs. 14-16).

The consistent findings around the stone pavement and the discovery of an area for sacrificial ceremonies suggests that this area was connected with the sacral sphere.

On the southern side of the stone pavement, ruins of two consecutive buildings were discovered in recent years. They are Buildings 3 and 4 probably used during the ritual ceremonies as shown by the number of sacrificial pits discovered in there.

Building 4 was burned and its foundation stones were taken for the construction of new structures; also modern agricultural activities caused the ruin of this building. Excavating under the level of the building, remains of another structure were discovered. The older ruins of this structure, called Building 3. It included sacred rooms with altars and water basins. In the initial part of purification ritual, the ritual patron – perhaps the great king – had to follow a precise procedure to perform the sacrifice as washing himself and dressing up pure clothes.

Altars made of mud-bricks were discovered in both buildings. The altar in the northern building is smaller than the other. By this second one was also unearthed unique finding: a windowed and winged vessel, which probably stood over this altar.

A broad courtyard was recently discovered south of the stone pavement and in the west of the aforementioned buildings.

A cuneiform tablet from this court describes a purification ritual suggesting the function of the area.

Hittite cuneiform texts usually indicate that pits (Hurrian *api*) were dig into the ground and served as passages to the Underworld. Animals as bird, lambs, sheep, pigs were sacrificed in these pits. As, for example, it is shown in the following *excerpta*:

When they bring them (i.e. the statues) to the place of the pit, they put down the gods and they open nine pits. They immediately take a hoe and they dig. Then they take chest ornament and dig. Then, he takes a šatta-shovel and *huppara*-bucket, and he empties (them into the pit). Then he pours wine and oil (into the pit). He breaks the thick loaves and places them around (the pit) on one side and on the other side. Then he put a silver ladder and a silver chest ornament into the first pit. He put silver ear and hands on the ornament; they are in the bottom of the first pit. A scarf binds to the last one of the ears. When he finishes, he offers one bird to all for *enumašši* and *itkalzi*. He contaminates blood to the nine pits. Then nine birds and one lamp (exist) for nine pits. He offers nine birds and one lamp for *ambašši*. He puts one bird into the each pit; but he cuts the lamp in pieces and put them into the first pit.

#### 104 *AYGÜL SÜEL*

The unearthed sacrificial pits, animal bones, and especially the bird nails and beaks are the archaeological proofs of the existence of these sacrificial pits (Fig. 17).

They have different forms serving probably different purposes. Most of the samples in the southern part of the stone pavement are deep. These pits are sometimes dug as one, double, triple or even quadruple (Fig. 18). In the double pits, it is seen that, the partition is extremely low as 1-2 cm (see the contribution of M. Süel in this book).

As far as it is seen, some areas with pits are surrounded with walls which were built with the dry wall technique and loose brush. The placement of the walls during the building process seems quite unorganized. While there are small stones in many lower parts of the wall, some bigger rocks were placed on the top. Also the thickness of these walls is not in the same standard. Parts 60-70 cm thick and parts 30 cm thick are very close to each other. They are not always straight line. Probably the most interesting fact is that some parts of the wall turns around a sacrificial pit, some other come across to a pit stopping on one side and starting again from the other side (Fig. 19).

Bone materials appear in three different forms in Ağılönü sacrificial area (Fig. 20). In the ceremonies in which birds and sheep (goats, sheep, and mountain goats) were burned together. According to the ceremony of animal sacrifice, after that the blood of the animal was poured into the pit, its head was put also there. The body, instead, was eaten or it destroyed as soon as possible. The bones usually appear as jaw, tooth, rib, leg bone, arm born, and sometimes vertebras. It is interesting that the leg bones and the sacrum ones rarely were unearthed. This variety denies the idea that only certain parts of animals were used. For example, a whole piglet was found in a pit. The sacrifice of a pig is described in cuneiform texts as follows (KUB 32.115 ++ II 44-57):

The old woman takes a piglet, and holds it over them (the patients) and says «It grew fat with meadow and barley and as it wouldn't see the sky and it wouldn't see (other) pigs anymore, let the prayers not see the evil curses. They hand on the piglets (over) them. Then, they kill it and dig the ground; and they put down the piglet. Then they put down the sweet bread and pour wine. Then they fill (it with) soil».

In another text (KBo 11.14 III 5-9) is written:

As the sun is still bright, the old woman, who performs the ritual, takes a knife and put it into the bedroom. As night comes, a priest digs (a pit into) the ground under the latch by a [kn]ife. Then, she takes the piglet and slaughters it into the pit; she lets down its blood (to the pit). Then she puts (the dead body of) the piglet with offerings as flour, grain bread etc. and she takes a loaf; she breaks it and (also) puts [down] it (into the pit) [for] the gods. (Then she says as follows): 'May the anci[en]t gods (like) them! Eat and [drink] the bloo[d] of pig and flour, the grain and the breads! Eat until you are satisfied!' However, she makes libation three times into the pit […].

It seems that a piglet and a puppy were burned in Ritual of Tunnawi (CTH 409): «Then she (the Old Woman Tunnawi) lifts up the piglet to him (the owner of sacrifice), and she performs the incantation of the piglet. Then, she lifts up the puppy to him. She performs the incantation of the puppy».

It is surprising that the surrounded areas include only a few hearths. These hearths have either straight form or half-moon shape.

It is known from the texts that, during the ceremony before sacrificing the animal, officials hit the heads of the animals with a stone or with a similar tool in order to make them dizzy. During our research, inside and outside the pits we found softly trimmed or non-trimmed stones. Sometimes there are two, three or more stones inside the pit. In some pits, there are more stones than in others. Some of these stones were found in various depths; some of them were beneath the upper cover of the sacrificial pit. The absence of the rock in some pits is associated with the possibility of usage of a wooden knob. After that the animal was brought in front of the pit, priests hit its head with this stone, and the dizzy animal was easily sacrificed, and beheaded. After that the blood drained into the pit and sent to the gods, the head and the feet of the animal were put into the pit, and the remaining parts were cooked and eaten by the participants of the ceremony.

Except for votive vessels, and little bottles used to collect sacred water for the ceremony there are not many complete vessels. As known from the texts, clay wares were broken during the ceremony and some parts of them left in the pits as for example: «and later he breaks fully the *hane*šš*a* vessel».

Seals have been found inside some pits or outside of some pits. In addition, the spindle whorls have been unearthed inside and outside of the pits.

The bottoms of the pits are usually empty. In the ritual of Tunnawi the ceremony is described as following:

They sacrifice animals by paying attention to the feet and head of the animals. While they cook the meat, the soldiers dig pits, they dig them close to each other. As if the last pit has merged to the first pit. Meat cooks and everybody eats. At the same time, they prepare two sedge tents for the king and queen for the ceremonial bath. Later inside the pit dedicated to the Sun-Goddess, a new pit is opened and it becomes a small bedroom with tiny bed models (Fig. 21).

Later the pits are connected with red and blue strings. The tents for the ceremonies are placed in an area especially prepared for this purpose. This area could be the area C in Sapinuwa (Fig. 22).

It is known from the texts that the kings became gods after their death. The souls of the dead kings would reach the bright sky among other gods, after they pass through the dark passages of the Underworld. Especially the rituals performed on the 8th day of the funeral were performed to communicate with the Underworld. The reference to the pits and the bird sacrifices proves that this day was dedicated to the gods. The sacrifice of burning the birds in the 13th day of the funeral is performed not for the dead but for the place where the dead goes. The peace of the dead in this place will be provided with these sacrifices.

The whole area in the Ağılönü plain was used for the sacrificial ceremonies.

On the top of the hills of Ağılönü, starting from the South-east of the Stone pavement, we see several buildings around. Along the east side, starting from the North, building 3-4, buildings 5, 6,7, buildings 1-2 and in the southernmost buildings has been detected. Also there are other buildings which were unearthed in previous years. From the cuneiform texts we know that there are many buildings encircling the area of the sacrificial pits (Fig. 23).

As stated in the text KBo 10.45 (III 26) at the end of the ritual «they cover the top of the sacrificing pit». This corresponds to the archaeological findings so that we can suggest that this clay layer was made to hide the pits that were dirty after their use. This fact makes really difficult to find these pits (Fig. 24).

The relation of the birds with the Underworld is complicated. Maybe they were perceived as messengers which communicated with the gods of the Underworld. Many rituals contain Hurrian sacrificing terms. Generally, in these rituals bird and few sheep and lambs were burned. After that followed a libation.

Rituals from Ortaköy are parallel to some texts from Boğazköy. Together with birds, there are animals like sheep and lambs and several sacrificing terms like in the following text: «On the mount Ḫazzi for the god Ea he burns the birds as: tell (*tepi*), speak (*kuliḫi*) sent (*pa*šš*itḫi*) reach (*amumgiḫi*), come (*pudiḫi*), tell (*tepi*), speak (*kuliḫi*), sacrificing bread in the size of a *tarna*».

In another text:

And when the morning comes they do in the following way: from the water that they get in, they take the water for the day, and they wash the gods. […] and they, offer the birds as: for the bird *enuma*šš*i* and *ku*š*ur*š*i*, a bird for pull (*aganti*), a bird for š*adandi*, a bird for aggregation (*tari*), a bird for *dupurpuri*, intimation (*ḫiliḫi*) and pouring (*tabiḫi*) [ ], a bird for the pouring (*tabiḫi*), and two birds for *elmi* and *parni*, a bird for present (*da*š*i*), *ari* and *mutri*, a bird for *zupi*, present (*da*š*i*), for *upzuri*, a bird for poring (*tapti*) and pouring (*tabiḫi*), a bird for chair (*tapri*), hand (š*uni*) and present (*maganti*) a bird for tell (*ḫilammiḫi*), and a white lamb for *a*š*ap*š*i*, a white lamb mollify (*enuma*šš*i*), and for purity and cleanness (*itkalzi*).

People probably believed that, during the burning ceremony, together with the birds also dirtiness would disappear, while the remaining ashes would dissolve in the soil and become harmless.

The magician takes the sacrificial animal hanged on him, and he touches the animal slightly or spit on its mouth to transfer the dirtiness. After that, they bury it into the pit that is a gate to the underworld.

#### *4. Conclusions*

The fact that the 'mouth-cleaning' rites were prepared in Sapinuwa is one of the important results connected with the exacavations of the site and show its religious importance.

Hurrian *itkalzi*-rituals written in Sapinuwa by Sapinuwian priests served to purify a human being from his sins. The sacred water, which could be provided from seven or even nine water supplies, was indispensable for these cultic ceremonies.

Sapinuwa kept its religious importance until the end of Hittite Kingdom. Treaties and official acts were fulfilled under the testimony of Oath-Gods. The fact that both Oath-Gods of Sapinuwa are still mentioned in the period of Tutḫaliya IV, as the ritual remains in the city, clearly prove this suggestion.

#### 106 *AYGÜL SÜEL*

A number of religious texts point out that cultic ceremonies were performed in various towns such as Sapinuwa, Sapantallia and Sulupassi. Cultic festivals were also fulfilled for the Storm-Gods of some towns as Serissa.

As we concentrate to Ağılönü region, the stone pavement, the sacrificial pits, the sacred buildings surrounding a broad area including these sacrificial pits, the workshops serving for cultic ceremonies and the silos suggest that this broad area was probably used for ritualistic purpose. After the results of Ağılönü excavation, we can suggest that one of the most important and unique religious regions of Asia Minor was discovered.

Beside the aforementioned examples, official seals, cuneiform tablets and other artifacts suggest that the sacred Sapinuwa was one of the most significant ritual places in the Hittite period. The sacral area was probably selected for its specific features. Were the water supplies? Or did another reason, strictly connected with the sanctity of the place, existed for such a choice? After the Building 3 burned, is there any reason to build the new building over the older one? We believe that such questions will find answer in the near future by means of both archeological and textual research. At the moment, the best definition is the fact that Ağılönü was one of the most significant sacred areas of the ancient world. It enables us to solve unknown problems and know Hurrian-Hittite rituals were performed. Although in the Hittite world, strongly influenced by the Hurrian religious system, there are several sacred places that connected gods and humans, Sapinuwa was probably more important than all the others. As shown in the texts, the royal couple, Tutḫaliya III, the great king, and Taduḫepa, his wife, ruled from Sapinuwa all other regions subjected by Hittite Kingdom.

Another tangling question concerns the status of the Storm-God of the city. According to a religious text from Ortaköy there existed two Storm-Gods of Sapinuwa. Could the existence of these two gods be a sign of a Hurro-Hittite unity in Sapinuwa?

#### *Bibliography*

Collins B.J. 2002a, *A History of The Animal World in the Ancient Near East*, Leiden, Boston, Köln.


Goetze A. 1938, *The Hittite Ritual of Tunnawi*, New Haven.

Haas V., Wilhelm G. 1974, *Hurritische und Luwische Riten aus Kizzuwatna*, Kevelaer, Neukirchen, Vluyn.


Fig. 1

Figure 1. Tablet of the itkalzi-ritual. Figure 2. Building D in Ortaköy. Figure 3. Representation of an armed god at the entrance of Building D. Fig. 5

Figure 4. Ritual axe.

Figure 6. Golden stylus.

Figure 5. Weapons found in Building D.

Figure 7. Ritual jars. Figure 8. A panoramic view of the site. Figure 9-10. Clay Artifacts found in Ağılönü area excavations. Figure 11. Oven excavated in the west side trench of Ağılönü hill. Figure 12. Figure of a naked goddess.

Fig. 7

Fig. 8

Fig. 9

Fig. 11

Fig. 10

Figure 13. Stone pavement excavated in Ağılönü area. Figure 14. Clay lion head. Figure 15. fragmentary bull head in clay. Figure 16. Clay bull head. Figure 17. Panoramic view of sacrificial pits from the Ağılönü area. Figure 18-19. Sacrificial pits in Ağılönü area.


#### THE RELIGIOUS SIGNIFICANCE AND SACREDNESS 111

Fig. 20

Fig. 22

Fig. 23

Fig. 24

# THE SACRED CITY OF HITTITES: SAPINUWA. THE NEW EXCAVATIONS

*Mustafa Süel*

#### *Abstract*

This paper deals with the recent results of excavations in the site of Sapinuwa. Significant evidence has been exposed in the area of Ağılönü that has been interpreted as a Hittite ritual place.

\*\*\*

Since the beginning of the excavation in the important Hittite site of Sapinuwa our team concentrated its efforts in the lower city (Fig. 1) where large monumental buildings have been found (Figs. 2). As it is known, Sapinuwa was a capital city of the Hittite State for some time. On the hills that we consider the upper city there are both military and civil monumental buildings (Fig. 3).

This paper deals with the results of our work in the lower city and the interpretation concerning the city in the light of these results.

At the end of the excavations that we carried out, the acropolis of the lower city and the portions that lay down to the city walls as terraces have been exposed (Fig. 4).

Several buildings in the upper terrace have been found and protected. Certainly, the excavations of such a large city will go on for a long time. Geophysical studies have provided us with one of the most important supports to gain new information and to accelerate the excavation.

Each of the buildings in the upper terrace reaches about 5000-10000 m2 together with their yards and service buildings. The consequence was that we needed to remove a lot of excavated soil from the area to understand the overall plan of the city and the relations between the buildings. The geophysical studies provided us preliminary information about the presence of yards, structures used as streets between buildings, and a lot of other information. Such data can be understood clearly and in detail with later excavations.

Two noteworthy results of the recent research activities are the detection of a Hittite road and the discovery of the sacred area in Ağılönü.

Very recent studies have allowed to detect and plot on the map about 10 km of this ancient road. Crushed rock, river gravel, and a special mortar have been used to build this road, which is 2.30 m wide (Fig. 5).

It will be possible to understand better the physical structure of the road by further studies. The city of Sapinuwa lays in a long valley between the plains of Alaca and Amasya, which are connected by this road, which was important not only in terms of transportation but also in terms of military purposes. It was used to carry materials from the cities which were in relation with Sapinuwa, when it was the centre of the state; and by the patrols to control and track the movement of the enemy.

Documents from the Sapinuwa archives concern the concentration of soldiers and blind people employed for milling cereals from the nearby cities. This road must have been really important for Sapinuwa in order that it could play its role, as it is attested from the documents. We are also trying to determine where strongholds were located on this road.

The Ağılönü settlement has been object of research for more than ten years, because of its importance, having produced written documents and having been a centre of Hittite ritual activities. Ritualization has been always very significant for ancient people. It was thought that gods must be pleased by gifts for avoiding every kind of evil. The first step for pleasing gods was the cleanliness of the person both physically and spiritually. The Hurrian *itkalzi*-ritual texts found at Sapinuwa, a 'purification' ritual, is the demonstration of the importance of such practices.

#### 114 *MUSTAFA SÜEL*

The Ağılönü sacred place is one of the most significant areas in this respect, and the excavations have provided valuable information in relation to the Hittite ritual world.

Rituals for the gods were performed in other Hittite and Hurrian sacred places. Sapinuwa must have been, however, among the most important ones, because the Hurrian – Hittite perspective was concerned with every kind of sickness and disaster. To avoid from these disasters it was an obligation to purify and clean the soul and the physical being. One of the most important religious ceremonies of the Hittites is this *itkalzi* purification ritual.

Sacred activities can be detected both in this extended city and in other neighbour sacred areas. Still, Ağılönü must be the most important of these areas. Its sacred building is more than thirty thousand m2 large, and it has been explored for more than ten years. Until quite recently, our information related to this subject was obtained only from the Boğazköy texts. Now, Sapinuwa provides really significant outcomes not only because of its texts, but also for its archaeological data.

In 2012, a structure, identified as a large building with a foundation of 60 cm, has been encountered in the north trench (Fig. 6).

By looking at the successive renovations of the floor, we can say that, this structure has been used for a long period. At the present moment, it is not yet possible to determine the plan of the building. By examining the occurrence of kitchenware, ash pit, and grinding stone we can accept that this place was used as an atelier (to Figs. 7-8). In the east of this trench, combining the outcomes of the excavations with the geophysical studies, a 160cm diameter architectural structure has been discovered at first. It has a circular plan and it was built with a dry rubble wall technique. It is probably an oven. Further investigation around it, in the north-eastern direction, a 180 cm diameter pit has been revealed just next to it, while at the north-western corner of a building has been discovered. We can presume to have reached an early level since this structure is 180 cm deeper than the structure we reached in the north trench (70 cm deep) (Fig. 9).

Considering the anomalies showed by the geophysical studies, investigations in 2013 have been carried out in the area at the east and northeast of the wide, plain area that we excavated the previous year.

In this area, in some levels, evidence of an extensive fire has been recorded. We started the excavation where the traces of fire are most extensive, in connection with two buildings which are located 100m west of a group of buildings which consist of ateliers placed around a street that connects the buildings 1-2. We called these operation 'south' and 'north' trenches.

There is 10 m between the two trenches. The south trench is 15x15 m. Here a portion of the settlement which consists of small and independent units has been observed. At the south eastern corner of it, positioned at the centre of the trench is an oven of 1 m of diameter. The oven with a conical body has been used for a long time. From the inside of this room, also from just in front of the eastern wall, many complete or fragmentary pots and copper slags have been found (Fig. 10). Probably this is a workshop for processing metals. However neither manufactured nor semi-manufactured goods nor their moulds have been discovered.

Just outside of the door opening, on the southern side, a possibly porch was discovered. We have to pay attention to the fact that this oven was placed outside the atelier. Open-air places would be useful for special activities, potentially dangerous for the health of the metalworker, such as the moulding of arsenic material or moulding lead. On the other hand, we can see similar ovens at the outside of the eastern corner of this building.

At the surroundings of these ovens we came across plenty of clinkers and melted fragments of pots. This atelier and the constructions around it are of low quality, sort of slipshod buildings with complex grid system, muddy mortar and ballast.

Probably the foundations are of mudbricks. Yet, neither these materials nor the roof and strut rods were preserved. The reason for this is the fire event. As far as it is seen, the reason of the destruction of these buildings and the other structures was an earthquake. In the studies that we carried out with the geomorphologists and geophysicists, strong evidences of such an event have been observed. As it is known, some geologists are in favour of the existence of another Çekerek-Kırıkkale fault line in addition to the North Anatolian fault line. The mentioned fault line waits still for a long period and become active from time to time and creates earthquakes in the high intensity. The debates about the existence of this fault has gained importance with the discoveries in Sapinuwa. The destruction from the east direction is really powerful. It is so powerful that it shifted the foundations, besides collapsing the construction. While the end of the buildings with strong timber ended up with fire, the small scale structures ended up with destruction. The ground of Sapinuwa, characterized by cystic and clay soil, don't liquefy, but it sinks instead. As it is seen, such a huge city ended up with destruction in a snap.

Another study area is the north trench, opened where the geophysical studies indicated intensive anomalies caused by fire. An extensive amount of cinder was found in the course of our investigations. At first this evidence was considered as pertinent to an oven. But the finding of a heap which was composed by 40x40 cm mudbricks with no stone foundations produced further evidence useful for interpreting this context. According to the excavations carried out in Ağılönü up to now, the fire which baked mud bricks in such a powerful way was the fire in buildings 3, 5 and 6. In these structures, the mud bricks and stuccos nearly become stone because of the effect of fire. There is no doubt that the mud bricks that we encountered here come from these buildings.

The widening of the excavations produced a great amount of kitchenware around and under this heap and a stonewall at the north side of it. This well constructed wall has no foundations. Orientated east-west, it continues in a south direction delimiting a square place. The most interesting finding of the 2013 season was found here. The floor, burned down by a strong fire, is very red in colour. There is no trace from burned woods. At the same time, the fire expanded in a large area (Fig. 11).

During the cleaning of the foundation, at the level of the floor, a complete tablet has been found. The interesting thing is that this tablet was not baked originally. Probably it was moved here after it was written and left inside during the destruction..

A vessel found in the same building (Fig. 12) is really important for us for knowing better about the Hittite ritual world. This vessel (Fig. 13), which will be probably completed after the detailed examination of the other fragments collected in the area, is an item that we know from the texts. It is a lamp and its Hittite name is šašanna. According to the literature, this is the second example of this category of vessels which has been recovered, the other one being also discovered in Ortaköy. Another object found in the same trench is a tripod stone altar. It was just near the stone paved road at the south-western corner of our trench. It is made from a rock similar to the andesite, and it is broken.

A stone paved road has been discovered at the south-western corner of the trench in the final days of excavations. The road is oriented south-east/north-west. It is too early to determine its function, but it seems to me that this can be connected to the entrance of the sacrificial area.

At the north-western corner of our trench there is a structure in form of a semi-circle that joined from outside the western margin of the stone wall. This place, 6 m2 large, is directly related to a room where is the mud-brick heap, and it has a door opening.

The floor was totally covered by ceramic fragments (Fig. 14). Having removed these fragments, a really interesting object came to light (Fig. 15). It is a stone mould, which was broken at its own place, found upside down under a broken pot. It is made of a material named wrongly as Babylonian glass in the literature, which we consider a specially prepared mould paste. There is an L shaped border line, without any figurative work. On the front, at the bottom, a lion in a resting position can be seen. Right above the lion, a goddess stands holding a fabric; in the centre of the composition an altar is in front of the goddess. Behind the goddess, at both of her sides, there is a god. The hands of the goddess, and those of the two gods at the both sides of the altar, rise slightly up by holding a sun disk.

Considering the results of the excavations in the Ağılönü area, we can state that a sacrifice area has been found, which starts from the stone pavement and continues to the south. As it is known, the texts mentions this kind of structures by referring to them with the determinative Ė. Yet until today none of these kinds of structures has been found. The first ones have been discovered in Ağılönü, south of the structures that we called 'Buildings 1-2'. These buildings are ateliers connected each other. They are placed at the side of a paved road.

North of these buildings, Building 7, a *pisé* construction, has been revealed.

In the north of these buildings, lining them to the north, Buildings 5, 6 and 3 and Building 4, which was built after the fire in Building 3, have been discovered. Buildings 7, 5-6 and 3-4 have religious functions.

Excavations at the west of the wide Ağılönü plain have brought to light more workshops. Here, ovens, big ash pits, broken findings which had been thrown into these pits provide us information. Recent investigations started last year in the south portion ensured us that the sacrificial pits end in this direction, and here it opens instead an area with ateliers (Fig. 16).

At the end of our researches to date, we were able to identify at least 3 architectural levels at the Ağılönü sacred place. The architecture of the latest period is extremely poor. This must be related to the poverty of people and the depopulation in the city. Reused material and river gravel is the evidence that can be observed, which do not present any architectural special feature.

The settlement, characterized by two layers, was destroyed because of a powerful earthquake. The city burned down and was destroyed. Yet it is hard to say that it was this earthquake to damage the stone pavement which is the most spectacular structure of the settlement.

#### 116 *MUSTAFA SÜEL*

This stone pavement, about 1700 m2 in size, is one of the biggest monumental structures discovered. The relation with the ritual activity is a fact. It was built with enormous effort and an outstanding technique, which we don't meet in any other place. Certainly, it must have been an important place.

Significant data has been obtained in relation to the settlement of the Ağılönü ritual place and the characterization of the structures. A lot of information that we know from the texts has been identified archaeologically and this identification continues. This stone pavement, one of the most important elements of this sacred hill, and the sacrificial pits, with the structures surrounding them related to ritual processes, represent different portions of this unique site. Some portion of these buildings must have been dedicated to ritual ceremonies; other portions were service structures for other daily activities.

As it is known, some kinds of materials must be used in the purification rituals, and Sapinuwa had to be the place where these materials were produced. The water used for the *itkalzi*-ritual came from here. Our hypothesis is that the pots used for the seven deities, the water related to the purification rituals, the materials used as amulets, and the sacrificial offerings for ceremonies, were also produced in this place.

Obviously, we need other and more in deep research for better understanding the significance of the Ağılönü area. We are trying to define the exact function of the stone pavement, the whole list of the materials here manufactured, the kind of production together with the ovens found in the ateliers, and the access between the acropolis of Sapinuwa and the settlement of Ağılönü.

The tablets discovered in Sapinuwa, seals, moulds, the magnificent architectural works show that this sacred city was a very important ritual centre. It is hard to say if there was any perception of sacredness related to this area before the founding of the city. The creation here of the Hittite version of the *itkalzi* texts and the necessity to perform this ritual also in Hattusa, even the definition of this city as the most purified one of the whole country, the choosing of Ağılönü as the most purified part of this city, are all substantial reasons denoting the importance of this settlement. Probably the closeness to water is at the base of the sacredness of the city.

Today, Ortaköy has a limited water supply, but sources must have been abundant in the past, when Sapinuwa emerged as an important centre during the 14th century B.C. All of these, and other similar questions, show us that there is a lot of research to do.

#### *Bibliography*


Figure 1. The site of Ortaköy/Sapinuwa.

Figure 2. The monumental buildings in the lower city of Sapinuwa.

Figures 3 and 4. Structures and buildings found in Sapinuwa.

Figure 5. The traces of an ancient road.

Figure 6. The portion of a building identified in the north trench.

Figure 7-8. Fragmentary findings.

Figure 9. The early level of the building found in the north trench.

Figure 10. A room in the south trench.

Figure 11. The traces of a fire in a room of the north trench.

Figure 12. View of the building in the north trench.

#### THE SACRED CITY OF HITTITES: SAPINUWA 121

Figure 13. A vessel found in the north trench. Figure 14. Fragmentary vessels and sherds on the floor. Figure 15. A stone mould found on the floor. Figure 16. Structures and buildings of the Ağılönü Area.

Fig. 15

# BUILDING RITUALS ATTESTED AT THE BRONZE AGE SETTLEMENT OF SALAT TEPE, DEMONSTRATING LUVIAN, HURRIAN AND HITTITE RITUALS IN THE UPPER TIGRIS REGION

*Ayşe Tuba Ökse*

#### *Abstract*

The excavations undertaken at Salat Tepe on the Upper Tigris River revealed six building levels dating to the 23rd-16th centuries BC, providing several contexts interpreted as evidence for ritual practices. The buildings were reused in later levels, and some parts of the former buildings were filled with spoil and mud before the construction of a later one. Nearly all levels were damaged by fire and earthquake, and some buildings seem to have been used for another purpose in a later phase, such as a domestic building as a grave or a cultic one as a domestic building. In these cases rituals were performed by offering figurines and slaughtered animals. For 'sealing' a building offerings were placed into small pits dug along the new foundations, or put into a bowl placed upside down on the fill of the former building. During erection of a new building miniature vessels or clay figurines were laid into the foundations, or upright placed bowls containing offerings were placed among the foundation stones. Most of the offerings are new born piglets. Pieces of large cattle were also attested in a number of ritual deposits. Small cattle had been laid in a pit as a whole skeleton. In a few deposits pieces of deer and dog skeletons are uncovered.

Similar ritual practices are stated in a wide region in the Near East. On the other hand, the Middle Bronze Age inhabitants of the Upper Tigris region are presumably Hurrians; Tunip-Tešup of Tikuanni, probably a Hurrian kingdom located in the Upper Tigris Region, had been recorded as a vassal of the Old Hittite King Ḫattusili I. The historical relationship correlates with the transfer of ritual practices originating from the native Hurrian culture to the Hittites; so, the ritual activities performed at Salat Tepe might have been originated from Hurrian traditions described in Hittite texts.

\*\*\*

#### *1. Introduction*

Salat Tepe is one of the ancient settlements in the water reservoir area of the Ilısu Dam, on the northern bank of the Salat river, a tributary of the Tigris River with access from the North. The excavations were undertaken in advance of the construction of the Ilısu Dam on the Upper Tigris River at Salat Tepe near Bismil/Diyarbakır in south-eastern Turkey. Salat Tepe is on the northern bank of the Salat River, *c*. 5 km to the north of the Tigris valley, *c.* 90 km to the east of Diyarbakır.

The excavations brought out five periods defined by gaps and different structures (Ökse *in press*). The earliest levels achieved in the southern trenches of the southern slope present several architectural levels dating to the Chalcolithic period and the Early Bronze Age I. Following a gap of ca. two centuries, the mound had been re-occupied in the Late Akkadian Period. The stratigraphical sequence on the mound summit provides us to produce a continuous cultural sketch of the time from Early Bronze Age IV to the Late Bronze Age I. The mound had been once more abandoned in the Middle Assyrian Period, and housed temporary dwellings of nomadic communities during the Iron Age and the Middle Age.

The Bronze Age levels provide several contexts interpreted as evidence for ritual practices performed during erecting, converting and sealing the buildings (Fig. 1). Levels 6 and 5 had been used in

#### 124 *AYŞE TUBA ÖKSE*

the Early Bronze Age IV (Period IIA). Radiocarbon analyses give a time span from 2150 to 2050 BC for level 5 (Yoneda 2012). The walls belonging to this level were reused in the earlier phase of Level 4; the new buildings were built in the later phase, on the debris of the former buildings. Level 4a represents the transition from the Early to the Middle Bronze Age, and 4b the earlier phase of the Middle Bronze Age (Period IIB). All these settlements are marked with traces of damages caused by earthquake and fire; moreover, rituals had been performed for converting domiciles to graves. Level 3 had been strongly levelled towards the end of the 18th century BC in order to establish a plain surface as substratum for Level 2 (Period IIC). According to radiocarbon analyses, the building complex at Level 2 suffered an earthquake and fire in the beginning of the 16th century BC (Ökse, Görmüş 2006: 142; Ökse *et al*. 2009; 2010), and several rituals had also been practiced on its ruins for sealing this damage.

#### *2. Foundation Rituals*

#### *2.1 Sacrificial Pits under Foundations*

The building at Level 6 suffered an earthquake and fire, according to the deformed mud brick walls with burned surfaces. The building had been filled as a substratum for building activities in Level 5. The central summit buildings of Level 5 were renewed and higher floors were established on the debris. New mud brick walls in various colours resting on pebble foundations were built on top of the former walls. Beneath the stone foundations of a wall with a ritual pit containing bones of a piglet has been found (Fig. 2). A similar foundation deposit of a newborn pig has been observed in Building G at the *piazza* in Hirbemerdon, located opposite to Salat Tepe, at the southern bank of the Tigris River (Laneri 2008; 2011; 2014: 122-123). Along the northern walls of a building in Level 5 a series of pits under the foundation level contain ash and animal bones; from east to west, the chin of a piglet, another chin, a piece of the vertebra of a lamb, a piece of a chin and bones together with burned wood remains. Similar sacrificial pits have been attested also under some walls of Level 4.

#### *2.2 Sacrificion in Bowls Placed Upright among Foundations*

A bowl filled with pieces of bones belonging to a piglet was placed upright among the stones of foundations of a wall in Level 4 (Fig. 3). The bones show no traces of burning, suggesting that the piglet had been placed in the bowl raw. A similar foundation deposit at Hirbemerdon Area B consists of a cache of one RBWW bowl and one Dark Rimmed Orange Bowl (D'Agostino *et al*. 2008: fig. 8b).

#### *2.3 Figurines under Foundations*

Level 5 consists of two or three-roomed domestic houses, separated by narrow paths from each other. Among the mud layers below several walls, pieces of baked clay figurines of quadruples and miniature cups have been uncovered. Some pieces resemble model horns of large cattle. The mound summit had been elevated and enlarged by high mud fills and mud brick layers supported by mud brick walls in Level 4. The architecture resembles the arrangement at Level 5. The paths are wider than the preceding level, and the new buildings were erected on top of the fills. Among the mud fills and in ash deposits on these fills as well as among the foundations of some walls pieces of baked clay figurines are collected. Deposit of various objects in the foundations of buildings for durability and prosperity of the inhabitants are widely practiced in Anatolia and the Near East. These findings deposited under foundations at Salat Tepe may also be remnants of a building ritual.

Under a mud brick wall, four baked clay objects are uncovered (Fig. 4). Two human figurines seem to have belonged to a male and a female. The male has a pointed hut and the female a flat one with folded edges. The breasts of the female figure are depicted by rounded knobs. The other objects are a drum-shaped miniature altar and a miniature bowl. This group might have been put into the foundations of a house as a sorcery ritual, in order to provide a good matrimonial order and fertility for the pair living in this house.

A similar practice has been observed at Tell Selenkahiye. Baked clay figurines – females and males wearing pointed caps – were broken and buried under niches, doorsills, ovens and walls of phase IIIhouses, before new constructions were built over them (Van Loon 1979: 100, 102-103; 2001: 6.343; Cooper 2006: 117; Peyronel 2008: 795). The bronze figurines found in the foundations of a Middle Bronze Age II building at Oylum Höyük (Engin 2011: 19) seems to have been placed to the foundations in the same manner.

#### *3. Rituals for Converting Buildings*

#### *3.1 Converting a Domestic House to Grave-House*

Level 5 is represented by a settlement composed by small buildings subdivided by narrow paths into lots. The eastern and northern walls of some rooms were built from the inner face, resembling slope houses placed partly semi-subterranean. The houses are composed of 2-3 rooms. At two buildings, each one room has horse-shoe shaped hearths, jars, cooking pots or pithoi placed into the floors, determining separate kitchens belonging to each house.

The walls belonging to Level 5 were reused in Level 4, and some parts of the former building were filled with mud brick spoil and mud for the construction of a later building. One of these domestic structures with two rooms, one with a hearth and a pot in the floor, had been rebuilt with stone foundations in Level 4a. The southern room became access to the northern room by means of a doorway, and the northern room a storing unit in the western half (Fig. 5).

In phase 4b the door had been closed with mud bricks, and long bones of a young deer were placed at the foundation. The legs of a double-hoofed animal had been placed to a pit under the foundation of the northern wall, and on the mud fill covering these bones, a white limestone in form of a bull-head had been placed (Fig. 6). Two burials laid into the northern room consist of an infant in a cooking pot and an adult female in a pit covered by a *larnax*. The female seems to have been placed to the room after the infant. The southern room had been filled with mud and small pits had been dug into the mud fill; one containing horns and others ritual objects such as miniature bowls, miniature stone axe and broken pieces of baked clay figurines (Ökse *et al*. *in press*: fig. 1b-c) (Fig. 7).

It might have considered that these individuals were inhabitants of this house, and after the child, also its mother might have been buried to their home. To the east of the building a room furnished by a U-shaped mud bank at the southern half contains several animal bones. A plain tripod portable altar recovered on the northern part the mud platform along the western wall (Fig. 8) points to the practice of ritual ceremonies related to the graves. Further ritual objects are found in the vicinity; a cubic terracotta altar and piece of a clay votive plaque with a spout protruding out, similar to those uncovered at Hirbemerdon (Laneri 2014: 123-124, fig. 4) point to the existence of a ritual behaviour around the building.

The practice of converting domestic houses to tomb-houses, especially for infants and children, is attested in Anatolia since the beginning of the 2nd millennium BC. Several burials under floors of domestic buildings are uncovered in Central Anatolian sites such as Kültepe (Özgüç 1950: 55; 1959: 11-12, 39), Alişar (Schmidt 1932: 88) and Boğazköy, in Tell Atchana in the Amuq Plain (Woolley 1955: fig. 62, 65, 68-69), in Lidar Höyük in the Middle Euphrates region (Akyurt 1998: 91-92, note 725) as well as in Syria (Akkermans, Schwartz 2003: 322) and in the Upper Habur region (Valentini 2011: 271, 277).

#### *4. Rituals for Sealing Buildings*

#### *4.1 Closing Doors*

Level 6 is represented by a building associated with open places on mud brick terraces on the western slope of the hilltop (Fig. 9). Small shallow pits on these terraces were probably used for placing vessels. The path leading to the main entrance from the southern terrace had been marked with tiny rows of pebble stones at each side. The door with a doorjamb at the inner part seems to have been a small door with a single wing fastened at the eastern wall of the entrance. The door leads to an entrance hall paved with medium sized pebble stones. To the east, another door leads to a 'libation room' associated with the terrace. A jar base placed into its mud brick paved and white plastered floor was probably used as a vessel stand.

Storage units to the west of the entrance hall could be reached either by a door from the west. At the entrance four mud steps lead to the stone paved floor. The room is heavily burned and several crushed vessels were found on the floor. A double framed door at the eastern wall leads to a narrow passage

#### 126 *AYŞE TUBA ÖKSE*

having access to the central part of the building by means of a stone stairway. The northern part is composed by three rooms with access to each other by double framed doors. The small room to the west is a heavily burned storage, similar to the previous one. The room has access to the second room to the east. This room is furnished by high clay platforms attached to its walls, most probably used as offering tables. Several vessels were found crashed on the floor paved with large pebble stones. The room has access to the north by a double framed door with a door-jamb. This door had been closed by mud bricks on an offering pit containing bones of a lamb, determining a ritual during closing. The door leading to the hall is preserved at the eastern wall.

The hall is the largest unit of the building. At the western end of its southern wall a double framed doorway for access to the entrance hall takes place. In a later phase the door was also closed with a short mud brick wall from the south, forming a niche. On the floor of the niche several bones of a big ruminant have been found (Fig. 10), determining a ritual for closing the door. The lower part of a terracotta human figurine found in the debris of the niche resemble the pillar-shaped standing figures with splaying, slightly concave, circular or oval bases suggesting standing figures in long garments are known from the EBA IV contexts (Early Jezirah 4-5) at Tell Selenkahiye (Liebowitz 1988: 4; Cooper 2006: 118; Van Loon 2001: 6.341), Tell Mardikh (Peyronel 2008: 790-792), Tell Mumbāqa Ekalte (Czichon, Werner 1998: Taf. 61: 200-202, 62: 209-213), in Tomb 302 at Jerablus Tahtani (Peltenburg *et al.*  1996: 12, fig. 13), in the 'votive offering deposit' at Hammam et-Turkman I (Van Loon 1988: 571-572) and in Akkadian houses at Tell Chuera (Moortgat 1962: 11, Abb. 7a-b).

A contemporary building with double and triple framed doors, niches and low podiums for offerings uncovered at Tell Beydar is also suggested to be a temple (Pfälzner 2011: 182, fig. 56).

#### *4.2 Sacrificing into Pits Dug into the Debris of Buildings*

A pig-rhyton recovered in a pit among the debris of Level 3 (Ökse *et al*. *in press*: fig. 1d) parallel to a rhyton from a 15th century BC context in Nuzi (Starr 1937: Pl. 103-7). Further pig rhytons are known from Tell Halawa, Hassek Höyük, Kültepe (Haas 1994: 536; Özgüç 2005: 172, fig. 19) and Tell Brak (Oates *et al*. 2001: 593.41, 47-50).

The heavily burned eastern unit building complex at Level 2 had been overlaid by a thick layer of mud and some shallow pits had been dug into the fill. Some of these pits contain only small pieces of animal bones, and into one pit, a lamb had been placed on its right side with its head pointing to the Northeast (Ökse *et al*. *in press*: fig. 2g). One of the pits contains broken pieces of terracotta bull figurines and small pieces of animal bones in an ashy fill (Ökse *et al*. *in press*: fig. 2i), and another one a quadrupled animal figurine with terracotta wheels (Ökse *et al*. *in press*: fig. 3j). These pits were then closed with a thick mud layer. On the ruins of the southern rooms, fragmentary floors paved with pebble stones and white plaster, feet of big ruminants had been placed (Albayrak Pekşen 2011; Ökse *et al*. *in press*: fig. 2f). There are cutting marks on several parts of the bones. At each foot a carinated bowl had been placed in upright position, resembling a libation ceremony combined with animal sacrifice.

#### *4.3 Sacrificing in Bowls Placed Face down on the Debris of Buildings*

In Trench K12 a bowl filled containing a piglet was placed face down on the debris of a building from Level 4, probably for sealing the building before erecting the new one. In another pit dug into a collapsed wall of Level 2, a bowl containing a broken female figurine formed by mud and a piglet was placed upside down. Also this offering aimed probably to seal the damage caused by earthquake and fire (Fig. 11).

#### *5. Discussion*

The Bronze Age inhabitants of Salat Tepe performed foundation rituals for new buildings, for converting domestic buildings to grave houses, and for sealing the buildings that suffered destructions by earthquakes and fire, before a new one has been erected, in order to get rid of disasters that ruined the settlements (Fig. 12).

The only ritual type attested in Level 6 consists of sacrificed animals on thresholds, pointing to the practice of rituals for closing doors. The ritual pits associated with Level 5 consist of animal bones in several pits dug under the foundations of walls and ritual objects placed under foundations, probably in order to protect the new settlement from the ritual contamination of the former building.

The archaeological data concerning with rituals in Level 4 show rich diversity. Animal sacrifice, offering ritual objects and sacrificed animals offered together with ritual objects had been used for foundation, converting and sealing rituals. The offerings had been placed into pits under foundations, in foundations or into bowls placed upright in foundations of a building. In this level the only sealing ritual is animal bones placed in a bowl laid face down on the ruins. The only ritual for converting a domestic house to a grave house is attested also in this level as parts of sacrificed animals placed into pits dug along the walls, and animal bones and broken figurines in pits dug into the fill of the former building.

One ritual type attested in Level 3 is a pig-rhyton in a pit dug into its ruins for sealing a building, and another ritual consist of laying a pebble stone idol under a foundation. Other ritual practices are not attested in this level, probably because of the strong levelling before the construction of the later building. The rituals in Level 2 seem to have aimed the sealing of the earthquake damage of the building complex. Either sacrificed animals or a piece of them had been laid together with ritual objects to pits dug into the ruins. In one case, animal bones and a figurine placed in a bowl laid face down on the ruins, and in other two cases, each one foreleg of big ruminants had been placed near upright placed bowls on the ruins.

The practice of animal sacrifice and offering of objects associated with rituals was common in the ancient Near East and Anatolia (Ebeling 1931: 3; Macqueen 1959: 173; Hoffner 1967: 389, 399; Steiner 1971: 265; Haas 1987-1990: 253-254; 1994: 908; Scurlock 1995: 1891; Ünal 1975-1976: 483; Collins 2002a: 225-226).

Hittite sorcerers practiced several rituals combined with digging sacrificial pits in houses, for getting rid of impurity, suffering, sorrow, curse, discord, perjury, homicide, or black magic (Otten 1961: 117; Erbaşı 2013: 408-412). Ritual objects and sacrificed animals were placed under foundations, and animals were offered after the erection of the buildings (Haas 1994: 48, 65-66, 256, 261, 266; Frantz-Szabó 2004; Beilke-Voigt 2007: 53-54; Görke 2011, lines 163-164, 223). Bilingual Hattian-Hittite ritual texts mention several offering practices and rituals associated with temples (Schuster 2002; Ünal 1988a: 97-98, 101; 1988b: 1470, 1473, 1477-1478; Süel, Soysal 2007); a foundation ritual describes sacrifices of sheep, libation and magical formulae spoken to ward off evil from the house. In Mesopotamia slaughtered sheep are used to purify the temple (Wiggermann 1992: 123; McCarthy 1969: 169). A text from Assur records two sheep slaughtered over the foundations of a new room (Ambos 2004: 71). Sheep had been offered to Gods by Hurrians at Kizzuwatna (Haas, Wilhelm 1974: 215-231).

The frequency of sacrificed piglets at Salat Tepe is a distinctive practice, attracting the attention to the Hittite (Friedrich 1959: 115; Süel 1985: 22, 157; Erbaşı 2013: 125) and Luwian-Hurrian (Ünal 1988: 56; 1996: 67, 76; Collins 2006: 174-176; Kümmel 1967: 152, note 9-10) belief considering dogs and pigs as dirty animals. According to the performance of the *walkui*-ritual, even eating or touching pork in dream had caused the people to be purified with rituals (Görke 2010: 208). Pigs were believed having contacts to the underworld and as being favourite dishes for Underworld Gods, so that the Gods would accept the offering rapidly, which rises the success of the ritual. Hittites sacrificed pigs during funerals (Herre, Röhrs 1958: 63-65; Otten 1958: 58 ff; Erbaşı 2013: 206-207, 280-281, 285) and Hurrian sorcerers used pigs as substitute for getting rid of badness and of demons causing psychological illnesses and for purifications from murder (Haas 2003: 139-140; Haas, Wilhelm 1974: 56; Collins 2006, 173, 177; Erbaşı 2013: 448).

Similar behaviour is also attested in Hirbemerdon Tepe IIIB, where a piglet had been placed in a bowl at the foundations of a ritual installation in the so called *Piazza* (Laneri 2011: 84; 2013: 24; Laneri *et al.* 2008: 184). A pit dug into the southern entrance of the Late Akkadian Palace at Tell Mozan (Urkesh) contains a great number of piglets and whelps as well as broken terracotta figurines of a pig and a woman (Collins 2004: 54-55; Kelly-Buccellati 2002: 131 ff; Lawler 2008: 50; Kıymet 2013: 297- 298). Similar ritual packages are also found in Hittite sites. Sacrificial pits with complete skeletons of piglets in Ortaköy (Süel 2010: 37, 49, res. 10; Erbaşı 2013: 286) and a bowl faced down on a piglet has been found in the rock-split chamber D in Yazılıkaya at Boğazköy (Hauptmann 1975: 64; Seeher 2011: 91; Kıymet 2013: 117).

These finds might have been associated to Hurrian and Hittite rituals (Kıymet 2013: 295-297). The Hurrian sorcerer Mastigga performed this ritual for calming disputed family members at Kizzuwatna (Reyhan 2003b: 165 ff) and Hantitassu for healing the restlessness of people at the city of Ḫurma (Ünal 1996: 22, 30; Collins 2002a: 20; Hoffner 1967: 390 ff). These rituals proceed with the scarification of a piglet into a pit where also drinks are poured, ash, seeds, puppies and butchered animals are placed (Wilhelm 1994: 74; Collins 2002a: 226; Collins 2004: 56; Erbaşı 2013: 375-376, 448). Hittites laid also models of pigs to the foundations in order to secure the wellbeing of the community (Collins 2006: 170).

#### 128 *AYŞE TUBA ÖKSE*

Similar to the piglets placed in bowls for the foundation and sealing rituals in Level 4 and 2 at Salat Tepe, skeletons of quadrupeds and birds associated with bowls and other findings, were also deposited under the doorway to the cella of the Parthian Temple of Inanna at Nippur and under a corridor in Ašurnasirpal's Northwest palace at Nimrud (Ellis 1968: 44-45, 130).

Several pits with broken pieces of baked clay figurines at Salat Tepe resemble the Hurrian rituals of the scapegoat; a goat symbolical loaded with impurity and devilry, is driven out of the settlement (Wilhelm 1994: 75; Reyhan 2003a: 132). This ritual is carried out with substitutes during the Hittite ritual 'Nakušši' (Ünal 1988: 83-84); impurity and devilry was loaded to figurines, and these were broken and buried, in order to keep evil out of the house (Haas 1987-1990: 247-248; 1994: 879, 886-887, 894-896; Maul 1994: 85-93; Scurlock 2006: 49-56).

The ritual activities performed at Salat Tepe is parallel to the magical rituals of the Hittite, Luwian and Hurrian traditions (Ünal 1996: 9; 1980: 477, 486; Haas 2003: 135-137, 141; Eliade 2000: 176). During the 18th-17th centuries BC (Wilhelm 1982, 12-19) Hurrian states ruled the region extending from the Diyala River to Cilicia, and in the 16th Century BC, the Old Hittite Kingdom enlarged its territory towards Hurrian settlements to the south of the Taurus range (Klengel 1965: 42; Wilhelm 1982: 29- 32). Tunip-Tešup, the king of Tikuanni, probably located in the Upper Tigris Region, was a vassal of Ḫattusili I (Wilhelm 1982: 21-22; Karg 1999: 275). The historical relationship of Hittites and Hurrians in the region verify with the Hurrian rituals practiced at Salat Tepe and Cilicia, and with the transfer of these practices to the Hittite-Luwian tradition.

#### *Acknowledgements*

The author wishes to express her thanks to Archaeo-zoologist Dr. Derya Silibolatlaz Baykara from the Yüzüncüyıl University (Van, Turkey) for the preliminary analysis of the animal bones mentioned in this paper, and to the Archaeologist Yusuf Akgün for preparing a list of animal bones used for rituals in the excavation data.

#### *Bibliography*


Haas V. 1994, *Geschichte der Hethitischen Religion*, Leiden, New York, Köln.



# **Building Rituals Attested at the Bronze Age Settlement of Salat Tepe, demonstrating Luvian, Hurrian and Hittite Rituals in the Upper Tigris Region**

Fig. 1. Table of Building Rituals Performed at Salat Tepe. **Figure 1: Table of Building Rituals Performed at Salat Tepe**

2400-2200 Late Akkad IIA: Level 6

Fig. 2. Ritual Pit L12/124/Ç containing bones of a piglet (L12/0297/H) beneath the foundations of Wall L12/116/D. Sealing Rituals

**Figure 12: Table of Ritual Types Attested at Salat Tepe**

Fig. 3. Bowl K13/0197/S/01 containing the skeleton of a piglet among the foundation stones of wall K13/42/D in Level 4. in bowls placed face down on ruins Level 4 and 2 near upright placed bowls on ruins Level 2

Fig. 4. The figurine group L11/0157/01-04 in the foundation of wall L11/047/D in Level 4.

1

Fig. 8

Fig. 5. A Structure in Level 4a with Rooms L12/092-093 and a Storage Unit L12/095/M.

Fig. 6. Sacrificial pit under the foundation of wall L12/087/D in Level 4b and a white limestone in form of a bull-head on the mud fill.

Fig. 7. Baked clay pig (L12/0203/P/04) and bull (L12/0203/P/03) figurines, horns and feet of quadruple figurines (L12/0215/P/02-10).

Fig. 8. Portable Altar L12/0286/P/01 from a Mud-Platform in Room L12/066/M in Level 4b.

Fig. 7

Fig. 9. The Building in Level 6 from the North. Fig. 10. The Bones of a Cattle Placed on the Threshold of Niche L11/143/D in Level 6. Fig. 11. Bowl J13/0001/P/03 containing a broken female figurine J13/0001/P/01 formed by mud and a piglet in placed upside down in Pit J13/0001/Ç in Level 2.

**Building Rituals Attested at the Bronze Age Settlement of Salat Tepe, demonstrating Luvian, Hurrian and Hittite Rituals in the Upper Tigris Region**

BC Period Salat Tepe Foundation Rituals Converting Rituals Sealing Rituals


1

Fig. 12. Table of Ritual Types Attested at Salat Tepe. **Figure 12: Table of Ritual Types Attested at Salat Tepe**

1550-1400 Early Mitanni IIC: Level 1 1700-1550 Hurri IIC: Level 2 1800-1700 Hurri IIB: Level 3 2000-1800 Hurri IIB: Level 4 2200-2000 Post Akkad IIA: Level 5 2400-2200 Late Akkad IIA: Level 6

**Figure 1: Table of Building Rituals Performed at Salat Tepe**

# RELIGION AND PROPAGANDA UNDER THE GREAT KINGS OF KARKEMIŠ

*Alessandra Gilibert*

#### *Abstract*

This paper focuses on monumental art decorating public gates at Karkemiš as a key to understanding the negotiation of political power in the period between 1200 and 950 BC. I argue that this kind of public art was first developed in Hittite Central Anatolia as a form of propaganda connected to state cults and formally bound to the centre of the Empire. After 1200, this art practice migrates south and is taken up by emerging polities seeking to perpetuate Hittite ideology. In the 12th century, Hittite-inspired public art is limited to the political milieu of the Great Kings of Karkemiš, with images centred on cult and kingship. In the course of the 11th century, the territorial influence of Karkemiš deflates and competing polities start their own Hittite-style public art projects. At Karkemiš, Hittite blueprints are forever abandoned in the first half of the 10th century, when public art shifts its focus from cult and kingship to the display of heroic force. I argue that this change of visual idiom is related to the rising political influence of a new class of governors, the Country Lords, and reflects the struggle of the Great Kings to negotiate a balance of power. By the end of the 10th century, the Country Lords reach full independence, the city's political identity changes radically, and public art morphs into something entirely different.

An identifying trait of the sacred landscape of Iron Age Luwian cities is the use of public space outside institutional buildings and holy precincts to commemorate and sometimes perform rituals and ceremonies, with much effort put into decorating architectural façades with monuments and inscriptions. In a number of important studies, and particularly in a seminal contribution on *The Gate and the City*, Stefania Mazzoni (1997) recognized Syro-Hittite city gates as places of religious and political significance, where monumental images were erected to illustrate and celebrate a characteristically urban ideology. This paper is a re-appraisal of the subject matter in the light of recent discoveries, with a specific focus on the history of Karkemiš from the dissolution of the Hittite Empire until the first half of the 10th century1 . The aim of this contribution is to examine public monumental art and its religious and political connotations in diachronic perspective, framing its development into the two major historical trends of the period in question: a) the geopolitical expansion and the subsequent slow but inexorable fragmentation of the Land of Karkemiš; and b) the rise and fall of its 'Great Kings', the self-proclaimed heirs of the Hittite Empire destined to be finally ousted by their own local governors, the city's 'Country-Lords'.

\*\*\*

#### *1. Public monuments and urban festivals in the Hittite Empire*

The practice of decorating urban gates with monumental art is rooted in Bronze Age Hittite ritual traditions, and linked particularly to the urban festivals that involved processions with sta-

<sup>1</sup> The present article benefited from the views and advice of Michele Cammarosano, Federico Giusfredi, Pavol Hnila, Federico Manuelli, Massimiliano Marazzi, Dirk P. Mielke, and Carlo Zaccagnini. To all of them, I owe my sincere thanks.

#### 138 *ALESSANDRA GILIBERT*

tions at the city gates and within the city (Mazzoni 1997; Görke 2008). Figurative stone reliefs such as those found at Büyükkale (Neve 1984; Orthmann 2008: 30) or, more recently, outside *Gebäude A* at Kayalıpınar (Müller-Karpe, Müller-Karpe 2009) and at the entrance of Building D at Ortaköy (Süel 2005; Süel, Süel 2013) indicate that Hittite rulers started decorating the façades of «buildings of special function» (Mielke 2011) with sculpted imagery related to the cult as early as the 15th century2 . In the course of the 14th and 13th centuries, monumental art expanded into public urban spaces, notably city gates (Schachner 2013: 539). When the city walls of the Hittite capital Ḫattusa were re-designed3 , three ceremonial city gates – the Lions' Gate, the Sphinx Gate, and the King's Gate – were decorated with, respectively, monumental portal lions, sphinxes, and the oversize image of a god4 . The elaborated figurative frieze at the Sphinx Gate of Alaca Höyük shows that, in the course of Hittite history5 , the decoration of urban gates with «the illustration of rituals in continuous narrative strips» (Mellink 1970: 18) acquired a significant position in the urban scenery. Written sources inform us that Hittite city gates were meeting places used by institutional power for public display and communication: at the gates, death sentences were carried out, prisoners exchanged, edicts read out aloud (Miller 2012). In short, the city gate was used by the Hittite king as «a forum to convey his authority» (Miller 2012: 679). Next to and intertwined with this 'political' value, city gates were also a place where religious rituals took place (Marazzi *in press* a; Marazzi *in press* b; Miller 2012; Dittmann 1999: 167-168; Voos 1983; Del Monte 1973). The figurative decoration of the gates at Ḫattusa and Alaca Höyük is directly linked to such rituals. Furthermore, the iconography chosen at Alaca Höyük to represent royal power and the royal persona expresses specific theological concepts and represents the ritual 'participation' of the king, through offering and libations, in the divine nature of the Sun-God and the Storm-God (van den Hout 1995). There are good reasons to assume that these gates were planned as places where religious celebrations were not only evoked in images, but actually performed. In favour of this view we may list a number of significant features: the presence of cup-marks for libations (Neve 1977-1978); the location of the decorated gates at nodal points of the cities' sacred landscape6 ; their stage-like architecture (particularly evident at the Sphinx Gate of Ḫattusa) (Schachner 2011: 158-164); the portrayed subjects (Hittite texts, for example, describes figures of sphinxes employed in the cult as «overseers of rituals» [Gilibert 2011a]); and finally, the presence of short Hieroglyphic Luwian inscriptions picked on or next monumental images. As argued by Marazzi, these 'signatures' relate to the transformative and protective power ascribed to the passage, and to the images that guarded it (Marazzi *in press* b).

The narrative reliefs at Alaca Höyük, with scenes of royal libations, ritual hunting, music, and games, fit well into our knowledge of Hittite festivals and suggest that at least some of the depicted events may have in fact taken place in the open spaces around the gate area (Mazzoni 2013: 472)7 . If this line of argument is correct, and considering the pivotal role played by the Hittite king and queen in the imagery at Alaca Höyük (Alexander 1989; Sievertsen 2008: 577), we may reasonably conclude that the practice of decorating urban gates with religious stone reliefs was initiated by Hittite rulers in the context of

<sup>2</sup> Figurative architectural reliefs were also known in Middle Bronze Age Syria: Kohlmeyer 2012: 65-66, and below. However, they appear to have been used in less accessible locations, for example inside temples, as in the case of a relief found reused in Aleppo. On the development of Hittite monumental reliefs, see De Martino 2010.

<sup>3</sup> A. Schachner (2011: 92-93, 159) favours a date for this event in the late 14th century, considering that the Lion's Gate was left unfinished and proposing a connection of this fact to the temporary transfer of the Hittite capital to the city of Tarḫuntassa initiated by King Muwattalli at the beginning of the 13th century. Arguments for an even earlier date are discussed in Simon 2011 and Schachner 2012.

<sup>4</sup> The Sphinx Gate seems to have been originally decorated with a number of further stone reliefs that were recovered only in fragments, which apparently included the standing figure of a male person: Bittel 1937: 7; Ussisshkin 1989: 486.

<sup>5</sup> For the dating of the Sphinx Gate at Alaca Höyük in the (late) 13th century BC, see in particular Neve 1994, with further literature, and, more recently, Orthmann 2002a, Sievertsen 2008, Taracha 2011. For an argument in favour of a date in the early 15th century, on the basis of parallels with Kayalıpınar, see Schachner 2012: 139.

<sup>6</sup> For the case of Ḫattusa, see Neve 1996: 17-21 and Schachner 2011: 158-164. For Alaca Höyük, see Neve 1994 and Sievertsen 2008: 575-577. Alaca Höyük was a small city near Ḫattusa, with a comparatively vast ceremonial and administrative centre, a fact that led to identify the site as a «cult city» (Mielke 2011) where important religious festivals took place, probably the ancient Arinna (Erkut 1992; Taracha 2011; although Popko 2000 proposes Zippalanda). The Sphinx Gate was the main entrance to the city centre, leading from a broad avenue (Neve speaks of a «processional way») to a urban square. From this square, access to the royal palace complex was granted.

<sup>7</sup> The development of open spaces on either side of city gates into ceremonial squares, which is a signature of the Luwian and Aramaean cities of Iron Age Syro-Anatolia (Mazzoni 2008; Gilibert 2012), may in fact have its origins in Hittite urban design (Matthiae 2008).

a state cult that involved rituals performed *in loco*<sup>8</sup> . Apparently, an important aspect of these rituals was the religious legitimation of the king, who, together with his family, presented himself in front of a wider public in his role as sponsor of religious festivals and pious executor of the cultic calendar.

As far as the extant evidence allows us to judge, the Hittite practice of decorating urban gates with monumental stone reliefs was geographically limited to Ḫattusa and its environs, that is, to the symbolic and religious centre of the Empire. In Northern Syria, Hittite rulers and high-ranking subordinate dignitaries initiated significant architectural projects, involving the decoration of temples with both typically 'Hittite' and more locally inspired sculptures, as the Late Bronze Age figurative evidence at Aleppo, 'Ain Dara, and Tell Atchana show (Yener, Dinçol, Peker 2014; Kohlmeyer 2012: 65-66; De Martino 2010: 94)9 . Fragments of a relief frieze found out of context indicate that architectural sculpture in Hittite style was also employed in building projects at Karkemiš10. However, none of these projects apparently involved the decoration of a urban gate, or of a comparable public location11. If we consider that, both in Mesopotamia and the Levant, urban gates were important, multi-functional, and ritually significant meeting places12, and that their architecture involved the employment of monumental stone orthostats at least since the early 2nd Millennium BC (Matthiae 2000: 177-178; Harmanşah 2007), the fact that Hittite rulers appear not to have exported the habit of decorating urban gates to Northern Syria may imply a strong link of this practice to religious cults that were formally bound to the hearth of the Empire in Central Anatolia. Perhaps, with the demise of the Empire, the ritual bounds that held this monumental habit anchored to Central Anatolia loosened up: if so, this may explain why the habit of decorating public gates migrated away from its roots to be transplanted into South-Eastern Anatolia.

#### *2. The imperial cosmos and the art of the 12th and 11th centuries BC*

Immediately following the collapse of the Hittite Empire, former Hittite provinces and the *araḫzena utnē*, the 'outer lands' of South-Eastern Anatolia, re-organized themselves into small-sized local polities under the hegemony of presumably two «rump states» (Harrison 2009; Weeden 2013) corresponding *grosso modo* to the former Hittite appanage kingdoms of Karkemiš and Tarḫuntassa (Starke 2002; Wittke *et al.* 2012: 32). The Kingdom of Karkemiš, risen to new power, is so far the most visible and the better understood among the two powers. The last Hittite viceroy of Karkemiš, Kuzi-Teššub, a descendant of Šuppiluliuma I, survived the collapse of the Empire and profited from the dissolution of central authority, perpetuating Hittite ideological habits and replicating Hittite forms of territorial control (Wazana 2001; Dodd 2007; Mora, D'Alfonso 2012: 393-395)13. He considerably expanded the

<sup>8</sup> For the dichotomy between 'state' vs. 'non-state' cults, each one with its own apparently mutually exclusive set of musical and athletic performances, see now Cammarosano 2014a: in particular 162-164, with further literature.

<sup>9</sup> As pointed out by S. Mazzoni (2013: 473), the quarries of Yesemek and Sikizlar suggest the existence of an (itinerant?) Late Bronze Age workshop of stone artists working with Hittite conventions and style in Southeastern Anatolia and Northern Syria. This 'school' may have later survived the collapse of the Empire and sold its services to new Iron Age patrons.

<sup>10</sup> Cf. Woolley 1952: Pls. 50b, 51a-b, 66b. For a recent discussion, see Aro 2013: 252-253. For evidence of an older, Middle Bronze Age sculptural tradition at Karkemiš, see Marchesi 2014a; Marchesi 2014b: fig. 4; Di Paolo 2006; Mazzoni 1972: 80, with ns. 1-3.

<sup>11</sup> In Northern Syria, a specific tradition for monumental decoration of entrances, most notably with portal lions, did exist, and went back at least to the Middle Bronze Age, but was limited to temple architecture: cf. e.g. the evidence of Temple P2 and Temple D at Ebla (Matthiae 1992).

<sup>12</sup> See, most recently, May 2013.

<sup>13</sup> The drift to independence and the imperial attitudes of the viceroy of Tarḫuntassa and, perhaps, of Karkemiš go back to the latter period of the Hittite Empire (Giorgieri, Mora 2010; De Martino 2010; cf. also Payne 2014: 150, n. 7). Attitudes and destinies of the Kingdoms of Tarḫuntassa and Karkemiš in the transition between Late Bronze Age and Iron Age differ. In the 13th century, Kurunta, King of Tarḫuntassa, let himself be represented on monuments and seal with the iconography and the title of 'Great King', apparently with the forced consent of the Great King of Hatti (De Martino 2010: 91; Giorgieri, Mora 2010: 144; Singer 1996); however, all material traces of the imperial *allure* of the Great Kings of Tarḫuntassa dissolve concurrently with the collapse of the Hittite Empire. Conversely, the 13th-century kings of Karkemiš appear never to openly compete with the Great Kings of Hatti. In fact, the contemporary iconography of the seals of princes and officials of Karkemiš, including a seal of Ini-Teshub, King of Karkemiš, incorporated the image of a long-robed figure with a winged-disk over his head, which has been convincingly interpreted as a materialization of the Hittite royal title 'My Sun' and thus a direct iconic recognition of the supremacy of the Hittite Great King (Giorgieri, Mora 2010: 142). Perhaps, the political stance of the Kings of Karkemiš was more solid and less conflict-ridden than that of their counterpart in Tarḫuntassa (De Martino 2010: 92; Mora 2004). And accordingly, after the collapse of the Hittite Empire, far from disappearing from the political scene, the Kings of Karkemiš expanded their influence and claimed the title of Great Kings (see discussion below).

dominion of the Land of Karkemiš to the North, taking control of regions west of the Euphrates up to the Land of Malatya, which had formerly been the easternmost stronghold of the Empire, seat of a Hittite governor (Liverani 2004)14. At the same time, as evidence indirectly suggests, Kuzi-Teššub took on the title of 'Great King' (Hieroglyphic Luwian MAGNUS.REX)15, formerly prerogative of the Hittite king (Hawkins 1988)16, and set the precedent for his successors, who continued to claim the same 'imperial' title for the following two centuries (here and hereafter, the reader is referred to Fig. 1 for a provisional overview of the rulers of Karkemiš)17. In order to enforce their dominion, the 12thcentury Great Kings of Karkemiš appear to have applied a double system of territorial control, directly inherited from the former Hittite administration: in some cases, they acted as overlords of vassal polities governed by local chiefs18; in other cases, territorial control was direct and entrusted to governors. Kuzi-Teššub installed governors with the title of «Country Lords» (REGIO.DOMINUS)19 both in the city of Malatya and in Karkemiš itself (Hawkins 1995b)20. The Country Lords had, in one case at least, family ties with the Great King: at Malatya, Kuzi-Teššub installed as REGIO.DOMINUS one of his sons, PUGNUS-*mili* (I)21. A necessary and crucial, if as yet still speculative corollary of the early politics of the Great Kings is that, at least in order to implement administrative decisions, at Karkemiš scribal schools and archival practices continued to exist across the Late Bronze Age/Iron Age ridge and beyond, although the production of cuneiform texts on clay tablets was apparently entirely dropped – in favour, we must assume, of the *tabulae ceratae* written in Luwian Hieroglyphs that will later feature so prominently in the self-display of the local élites. The written tradition of Karkemiš in the first two centuries of the Iron Age, still almost completely eluding us, is the missing link to understand the stream of tradition behind the politics of the Great Kings (Aro 2013: 246-248)22.

The system of territorial control devised by Kuzi-Teššub was destined to slowly but constantly disgregate into its elemental parts. Probably already by the second half of the 12th century, the Country

<sup>14</sup> The significant expansion to the North, however, may have been counterbalanced by a loss of control in the south-western parts of the former kingdom.

<sup>15</sup> Hawkins (1995: 27-28; but see the discussion in Houwink ten Cate 2007: 191-193) proposes *šalli*-*ḫaššu*- as the Hittite rendering of Luwian Hieroglyphic MAGNUS.REX (= Cuneiform LUGAL.GAL), and *ura*-*ḫantawati*- as its Luwian counterpart.

<sup>16</sup> The earliest text ascribing to Kuzi-Tešub the title of Great King are the rock inscriptions Gürün and Kotükale, composed by his 'grandson' Runtiya, ruler of Malaya in the latter part of the first half of the 12th century (Hawkins 2000: 295-301). The possibility must be considered that Kuzi-Tešub himself never employed the title, and that it was ascribed to him by later rulers claiming direct lineal descent only as a posthumous construct (Aro

2013: 256, n. 119; Payne 2014: 150). 17 In the epigraphic records of Karkemiš, the title of MAGNUS.REX is employed by, in descending chronological order, Tutḫaliya (Karkamiš A16c; Karkamiš frag. a/b), Sapaziti, and Ura-Tarḫunta (Karkamiš A4b, and now stele KH.11.O.400), who all ruled in the (second half of the) 11th century BC. We may add to these occurrences the 'Great King' *I*(*a*)+*ra*/*i*-TONITRUS mentioned in the Karahöyük (Elbistan) stele (Hawkins 2000: 288-295). The stele, authored by a local magnate controlling the plain of Elbistan on behalf of the Great King in question, is dated by Hawkins on epigraphic grounds to the early 12th century. The political ascription of *I*(*a*)+*ra*/*i*-TONITRUS, whose name may be read Ir-Teššub, Yarri-Tarḫunta, or Ini-Teššub, is still a matter of debate (Harmanşah 2011: 65-69; Freu, Mazoyer 2012: 37-38). However, there is a strong argument (Simon 2013) in favour of an identification of *I*(*a*)+*ra*/*i*-TONITRUS with a descendant of Kuzi-Teššub, either his successor or a king who reigned at some point in the 12th century. Simon (2013: 828) thinks it possible that *I*(*a*)+*ra*/*i*-TONITRUS may be identical with the «Ini-Teššub, King of the Land of Hatti» met by Tiglath-pileser I c. 1100 BC, as reported in the annalistic texts of the Assyrian king (RIMA 2, A.0.87.3, ll. 26-28; A.0.87.4, ll. 28-30).

<sup>18</sup> The stele Karahöyük is the earliest evidence for the post-collapse politics of indirect territorial control of 'Great Kings' (of Karkemiš?) over local rulers. Further, it has been recently argued that a silver bowl with a Luwian Hierogliphic inscription from Ankara mentioning a «Labarna» Tutḫaliya and a «king» *(Maza)-Karhuha* may date to the post-Hittite period (cf. the abstract of the current discussion in Weeden 2013: 7-8, with references). If this dating should prove correct, the bowl could give further evidence for the early Iron Age political system involving the coexistence of 'Great Kings' and vassal kings.

<sup>19</sup> The Luwian full reading is not established: J.D. Hawkins (2000: 96) discusses the problem and tentatively proposes REGIO = \**utniyasi*, «of the country».

<sup>20</sup> The title REGIO.DOMINUS was derived by the imperial administrative structures: Hittite rulers appointed «Country Lords», translated on seals by the cuneiform EN.KUR*TI*, as governors with civic and military functions (Giusfredi 2010: 97-101; Payne 2014: 150-151, with further literature. For a Hittite Country Lord at Alalakh and its political ties to Late Bronze Age Tell Afis, see Archi, Venturi 2012). 21 For relations in lineage between Great Kings and Country Lords, see Payne 2014: 151-152.

<sup>22</sup> On the Luwian scribal tradition in the Hittite period, and on the 13th-century «scriptorial and scribal leap» from Hittite to Luwian and from the cuneiform clay tablet to the hieroglyphic wooden board, I refer to M. Marazzi, particularly Marazzi 2002, with further literature; see also van den Hout 2006 and Waal 2011. On the use of Luwian hieroglyphs at Karkemiš in the Late Bronze Age, see the works of C. Mora, particularly Mora 1998 and Mora 2010. On scriptorial techniques, see Cammarosano 2014b.

Lords of Malatya ceased submitting to the overlordship of Karkemiš, and pursued independent politics23. Malatya was the first, and northernmost, region to secede from Karkemiš. By the end of the 11th century, an independent Kingdom of Palistin, with its royal city at Tell Tayinat in the Amuq, was rising to considerable power and influence over the territory corresponding to the Late Bronze Age Hittite apanage kingdom of Aleppo (Harrison 2010). In the course of time, and increasingly so during the 10th century, further vassal polities followed suit and declared independence24, while Aramaean élites wedged themselves more and more in the interstices of Karkemiš' fading 'imperial' power.

At Karkemiš itself, however, Great Kings and Country Lords coexisted at least until the first half of the 10th century, with both titles being inherited by paternal lineage (Hawkins, Peker 2014; Payne 2014). Diplomatic relationships and power balance between Great Kings and Country Lords, as well as foreign politics of the Land of Karkemiš after the collapse of the Hittite Empire, are still largely unknown. Assyrian sources report that, around 1100, Tiglath-Pileser I met «Ini-Teššub, king of the Land of Hatti», who must have been a king of Karkemiš (RIMA 2, A.0.87.3, ll. 26-28; A.0.87.4, ll. 28-30)25. At Karkemiš, the epigraphic record dating to the 12th and 11th centuries boils down to three fragments of monumental inscriptions, to be cautiously connected to the Great Kings of the (latter half of the) 11th century:

	- (1) Great King Tutḫaliyas Great King Hero of the Land of Karkemiš
	- (2) King of the Land […; (Hawkins 2000: 82; Giusfredi 2014: 485)26
	- (2) […*kar*-*ka*-*mi*-*s*]*à* REGIO REX I *pi*-*ia*-*si*-˹*li*˺ […
	- (1) [Great? ] King? Tutḫaliyas [Great? ] King Labarna? …
	- (2) … King of the Land of Karkemiš, Piyassilis'… (Hawkins 2000: 590-591)28.

<sup>23</sup> The first Malatyan ruler who incontrovertibly ascribes to himself a royal title, REX.\*462 («Potent(?) King»), is PUGNUS-*mili* II, grand-grand-grandson of Kuzi-Teššub: cf. *Relief K* of the *Porte des Lions*, i.e., Malatya 8 (Hawkins 2000: 309-310), and the stele Malatya 14 (Hawkins 2000: 313-314). PUGNUS-*mili* II is also the first Malatyan ruler to initiate a cycle of monumental reliefs at the citadel gate, on which he is portrayed as king: public art appears here to connect with a conscious act of beginning a new political era. It remains to be explained why PUGNUS-*mili*'s son Arnuwanti II designates his father «Country Lord» (a title that is not attested in the inscriptions of PUGNUS-*mili* II), but titles his grandfather Arnuwanti I «King» (REX). The textual and visual contents of the steles İspekçür and Darende by Arnuwanti I leave two possibilities open: either the first ruler of Malatya to become independent from Karkemiš was in fact Arnuwantis I, father of PUGNUS-*mili* II, or Arnuwantis II construed a false tradition. In both cases, for reasons unclear, Arnuwanti II played down the political role of his father. What we know with a reasonable degree of certainty is that Runtya, an older brother(?) of Arnuwanti I and the commissioner of the rock inscriptions Kötükale and Gürün, did not claim royal titles neither for himself nor for his father PUGNUS-*mili* I.

<sup>24</sup> As in the instructive case of Gurgum (Porter 2003; D'Alfonso 2012: 190, n. 24) and, poorly documented, Kummuh, the classical Commagene. Cf. also the emblematic case of Tell Ahmar/Masuwari, located on the east bank on the Euphrates, which may have raised to independence from Karkemiš in the late 10th century BC, fell under the rule of the Aramaean tribe of Bīt Adini in the mid-9th century, and was then conquered by the Assyrians (for a different interpretation, according to which Tell Ahmar fell under Assyrian control in the reign of Tiglath-Pileser I, was then conquered by the Arameans, who appropriated the Luwian monumental language of Karkemiš, and was finally re-conquered by the Assyrians, see Bunnens 2013).

<sup>25</sup> It is interesting to note that, although the Assyrian sources use the expression «King of the Land of Hatti», reflecting the survival of a perception of Karkemiš as a fulcrum of 'Hittite' presence, the rulers of Karkemiš themselves never claimed a geographical control over a «Land of Hatti». L. D'Alfonso (2013) points out this fact to invite caution in believing in the will of the early rulers of Karkemiš to actively appropriate the Hittite imperial past.

<sup>26</sup> The stele is decorated with a winged sun-disc surmounted by a crescent and a planet. It may be provisionally dated to the late 11th century on the ground of its palaeographic affinity to the similar steles erected around 1000 BC by the Country Lord Suhi I (KH.11.O.400; Dinçol *et al.* 2014) and by his son Arnu-*x* (Karkamiš A4b; Hawkins 2000: 80-82), for which see below. The 11th-century date is also supported by the iconography of the winged sun-disc, to be compared with the sun-disc on the relief B30b, the so-called 'sacrificial slab', found at the Water Gate and dated to the 11th century on iconographic and stylistic grounds (Mazzoni 1997: 316-317, see also below).

<sup>27</sup> As indicated by the slanted edge of the 'proper end' sketched in the squeeze of the British Museum (Meriggi 1975: Pl. XVI).

<sup>28</sup> The dating is given to the late 11th century on palaeographic grounds.

#### 142 *ALESSANDRA GILIBERT*

III a small inscribed fragment (the upper hind leg?) of a portal lion, mentioning an otherwise unknown «Ḫuwa-Šarruma, King of the land of Karkemiš» (Karkamiš A18d, now lost)29.

These fragments corroborate further the existence of Great Kings at Karkemiš30 and show their cultural debt to the Late Bronze Age administration of the city. Format and contents of the inscribed stelas remind of Hittite antecedents31. The use of the determinative REGIO for the definition of Karkemiš as a political entity, as in *karkamissa* («REGIO»), follows the same use in the period of the Hittite vice-regency. The royal onomastic in general, and in particular the occurrence of the name Piyassili in Karkamiš frag. a/b, show strong ties to the Hittite forefathers and may even point to the persistence of an ancestral memory going back to the very first Hittite king of Karkemiš, Piyassili, son of Šuppiluliuma I (Hawkins 2000: 590-591; Payne 2014: 153, n. 31).

Scant but crucial collateral information for Karkemiš in this period is given by the Hieroglyphic Luwian inscriptions left by Taita, King of Palistin, in the Temple of the Storm-God of Aleppo at some point during the 11th century32. In Aleppo 6, offerings are prescribed according to rank and status (§5-9). Highest status is given to kings (REX), who shall offer an ox and a sheep. There follows Princes, Country Lords, and «River-Country Lords» (FLUMEN.REGIO.DOMINUS), apparently all belonging to the same hierarchical level (prescribed offerings: a sheep). Everybody else is listed generically as 'inferior', and shall offer bread and little else33. The passage does not acknowledge the existence of 'Great Kings' (Taita himself is titled REX): it seems that, by the age of Taita, the title MAGNUS.REX had morphed into a simple variant of REX, linked perhaps exclusively to Karkemiš, and did not automatically imply superiority over other titles. From the inscription Aleppo 7 (§2), we infer that Taita, at a certain point, for reasons unclear, was able to enter a secluded place of significance (DOMUS *sa5 -sa5 -tà-ti*: a treasury? a granary?) within the walls of Karkemiš (Hawkins 2011: 49). The nature of this 'visit' is yet to be understood: the context seems to imply a singular event of plunder. The presence of Taita at Karkemiš and his disregard for 'Great Kings', combined with the dearth of local epigraphic and archaeological evidence, epitomize the loss of territorial hegemony and the increasing presence of political competitors that incurred to Karkemiš in the 11th century (Weeden 2013: 6, 10)34. However, the fact that Taita turned his presence in the city into literature, whatever nature and duration of this presence might have been, seems to reflect the continuing cultural aura that Karkemiš exerted on its neighbours.

Turning our focus on monumental art, the production of the 12th and 11th centuries (corresponding, in terms of the archaeological sequence defined by S. Mazzoni (2000), to the phase Iron Age IA-B) may be divided into two main sets: art produced for temples and temple precincts35, and art produced for public gates. Both sets reflect the same continuity with Hittite traditions highlighted by the textual sources. In this respect, the extraordinary artistic evidence from the temples of Aleppo and 'Ain Dara is emblematic. Both temples were kept in use and variously re-decorated throughout the 13th to the 9th centuries. In the late 11th century, Late Bronze Age images of sphinxes, lions, and *Mischwesen* were still downright copied, in the obvious attempt to replicate a style and iconography unmistakably bound to the Hittite Empire (Kohlmeyer 2008, 2012)36.

<sup>29</sup> J. D. Hawkins (2000: 83) proposes a date on palaeographic grounds in the 11th century, «somewhat older» than Karkamiš A16c. If this is correct, then the lion fragment of Huwa-Sarruwa would be the oldest Iron Age epigraphic record in Karkemiš.

<sup>30</sup> The fact that Karkamiš A18d titulates Huwa-Sarruma as «REX *karkamissa*(REGIO)» does not rule out the possibility that Huwa-Sarruma also bore the title MAGNUS.REX, which never appears specified by a genitive: I thank Jörg Klinger for pointing this out to me.

<sup>31</sup> Cf. the stele of the «Great King, Tabarna Tutḫaliya» (IV) on exhibit in the Boğazköy Müzesi (Neve 1992: 34, fig. 84-85).

<sup>32</sup> For these inscriptions and their dating in the 11th century, see Hawkins 2011.

<sup>33</sup> Cf. with the later inscription of Katuwas, Karkamiš A3, §19: «whether he is a REX or a REGIO.DOMINUS […]» (Giusfredi 2010: 99).

<sup>34</sup> See also Summers 2013: 316 for evidence from Tille Höyük inviting to rethink the political weight of Carchemish in this period.

<sup>35</sup> I include in this set the subset of inscribed steles, including four steles from the Kingdom of Malatya (Hawkins 2000, Malatya 13, Malatya 14, Darende, İspekçür) and a stele found at Samsat (Hawkins 2000, Samsat 1 – *pace*  Hawkins 2000: 333, the royal figure with long robe and lituus is indicative of a date to the 12th century). Although none of these steles has been found *in situ*, dedication to gods mentioned in their inscriptions and comparisons with cognate inscribed steles found within precincts (Karahöyük, Karkamiš A4b) suggest that this kind of artefact were destined to locations of limited access, probably temple courtyards.

<sup>36</sup> Pushing K. Kohlmeyer's line of argument further, we shall consider the possibility that the *Sockelreliefs E*, decorating the socle of the cella inside the temple of 'Ain Dara and representing 'mountain gods' and other

It appears that city gates were renovated and re-built at a faster pace than temples, and thus the early fates of public art are known only in outline, and mostly from spolia. Nonetheless, a corpus of gate sculptures reminiscent of Hittite antecedents indicates that, within the Karkamishean milieu, Luwian rulers took up, adapted and developed the Central Anatolian habit of decorating important urban gates with symbolic and religious imagery. So far, evidence for longer figurative cycles and, significantly, for short Luwian Hieroglyphic captions37, has been recorded only at Karkemiš itself and at Malatya. In the course of the 11th century, gate decoration spread also to less important sites, but was apparently limited to anepigraphic portal lions38. This 'trickle-down' of portal decoration may be yet another indication of the progressive erosion of Karkemiš's political hegemony. The loss of political strenght, together with an enduring cultural influence, may have opened up the chance and triggered the desire of emerging political subjects to emulate the city's most visible monuments. The earliest dated Iron Age example of a longer figurative cycle are the reliefs associated with king PUGNUS-*mili* (II) of Malatya, to be ascribed on genealogical, palaeographic, and artistic grounds to the second half of the 12th century39. Six fragmentary reliefs found re-used at the Water Gate of Karkemiš and several stray fragments may date to the same period (Woolley 1921: fig. 33; Pls. B. 28; B. 29a-b; B. 31a-b.)40. Two further relief blocks (Woolley 1921: Pls. B. 30a-b) and the fragment of a portal lion (Woolley 1921: fig. 32) from the Water Gate and further stray fragments41 are stylistically different and evidently later; they closely resemble the earlier Iron Age reliefs of 'Ain Dara and Aleppo (the portal lion is virtually identical to the portal lion found at 'Ain Dara: Fig. 2), and consequently are to be dated to the age of Taita (i.e., 11th century)42. A coeval addition of new reliefs to existing cycles is also recorded at Malatya,

*Mischwesen*, may be a 12th-century copy of a similar relief, the *Sockel G 1*, found out of context but certainly dating to the Hittite Empire period (see Mazzoni 2013, 473 for a Late Bronze Age date of the «1st style mountain gods»; for images, cf. Abū 'Assāf 1990).

<sup>37</sup> For Karkemiš, cf. the fragment Karkamiš A18d mentioned above.

<sup>38</sup> Cf. the 11th-century lions found at 'Ain Dara, Elbistan, and Tilmen Höyük. The lions of 'Ain Dara were apparently found re-used as paving slabs in a later gate (cf. the indications in Seirafi 1960 and Orthmann 1964). Their archaeological context of the lions has not been published in detail; for their analysis and dating on stylistic grounds, see Orthmann 1971, 58, 476, 'Ain Dara A/1-3; Kohlmeyer 2008: 124, n. 20, fig. 7.

<sup>39</sup> Delaporte 1940, *Reliefs B-K*. The reliefs B-E were found re-employed together with other *spoliae* at the *Porte des Lions*, a citadel gate dating to the 9th century. On chronology and stratigraphy at Malatya in the transition between Bronze and Iron Age, see Manuelli 2012. F. Manuelli also recently presented the results of his stratigraphic and architectural analysis of the *Porte de Lions* and an earlier Iron Age gate on which the *Porte de Lions* was superimposed, the so-called «Porta Schaeffer» belonging to the Malatya level IIIA2 (Manuelli 2012: fig. 1). Excavations of the IIIA2 context at the 'Porta Schaeffer' revealed a series of aniconic slabs and sculpted orthostats *in situ* belonging to Orthmann's 'Malatya III' style (Manuelli 2012: fig. 4). These orthostats were sealed by a destruction level dating to the end of the 11th or the beginning of the 10th century (Manuelli 2014). This approximate date is thus a terminus *ante quem* for the *spoliae* re-employed in the later *Porte de Lions*, including the PUGNUS-*mili* reliefs. For the dating of the *Porte des Lions* and its antecedents, see also Liverani 2011, 92-93. For the dating of PUGNUS-*mili* (II), see Hawkins, Peker 2014: Table 1. For the early dating of the PUGNUS-*mili* reliefs on stylistic and iconographic grounds, see Bunnens 2006: 51-52, summarizing the discussion and referring to the works of S. Mazzoni; Orthmann 2002b: 277; Brown 2008: 304-307.

<sup>40</sup> The Water Gate is a 2nd-millennium gate that underwent several re-modelling phases. All reliefs were found spoliated or re-employed in the masonry (for the find-spots of the reliefs, see Woolley 1921: 103-116). Several stray fragments belonging to the same period of the Water Gate reliefs were found reemployed as filling material in later context. These are: a) fragments of five reliefs representing anthitetical bull-men holding a palmette, one of them a corner-block (Woolley, Barnett 1952: fig. 70; Pl. B. 52b-f); b) the fragment of a somewhat different bull-men, holding a drooping plant (Woolley, Barnett 1952: Pl. B. 49a); the basalt relief of, presumably, a passing sphinx, found exhibited in the courtyard of the Storm-God Temple (Woolley, Barnett 1952: Pl. B. 48a); c) fragments of at least one orthostat representing a passing lion (Woolley, Barnett 1952: pt. 4, fig. 72).

<sup>41</sup> These include fragments of at least two basalt lions in the 'Ain Dara style (Woolley, Barnett 1952: 182-183, pts. 3, 5 and 6, figs. 71, 73); and a sphinx head found out of context, virtually identical to the 11th-century sphinxes at 'Ain Dara (Woolley, Barnett 1952: Pl. B. 67a).

<sup>42</sup> I follow A. Özyar and S. Mazzoni in dating the reliefs B. 30a-b on iconographic grounds to the earliest phase of Iron Age monumental art (Mazzoni 1997: 316-317; Özyar 1991: 29; see also Brown 2008: 325-327). For further antiquarian and iconographic clues speaking for an early date, consider the folding stool on B. 30b, closely related to Late Bronze Age images of furniture (Symington 1996: 129-132), and cf. B. 30a with the relief of a man leading a bull to sacrifice at the temple of Ain Dara (Abū 'Assāf 1990: 58-59, *Stele F3*; for the dating to the 11th century, see Kohlmeyer 2008). Taking further into consideration the stray finds listed above, it may very well be that the same 11th-century workshop of sculptors that operated at 'Ain Dara also operated at Karkemiš. Similar attestations of the same sculptural school at sites under different political is a recurring fact, evidently connected with the existence of wandering artisans: cf. the case of Tell Ahmar and Zincirli in the late-10th and 9th century, when local rulers hire the same sculptural workshop active in Karkemiš and let them produce replicas or shorter versions of the Karkamišean figurative cycles.

where, during the 11th century, a second, and perhaps a third set of reliefs with images of *Mischwesen*  were added to the PUGNUS-*mili* cycle43. In short, both at Karkemiš and Malatya changes in style allow us to identify two main sets of monumental art: a set dating to the 12th century and a set dating to the 11th century. At Karkemiš, this latter set corresponds to the 'Ain Dara-style reliefs. As we shall see below, it is tempting to imagine that this set of reliefs mark the renovation of the Water Gate after Taita of Palistin was able to force his army into the city.

The interpretation of these earliest Iron Age reliefs is hampered by the fact that they were all found re-used as *spolia* in later versions of the gates where they originally stood. This fact gives us an impression of how valued the reliefs continued to be through time, but also leaves us with nothing but the *disiecta membra* of the former visual message. On the basis of iconography, we can identify three main groups of images: portal lions (Fig. 2), which apparently grew into a sort of *conditio sine qua non* for decorated gates; representations of the king worshipping the gods (Fig. 3); and images of passing animals and *Mischwesen* (Fig. 4)44. Even if specific emblematic associations elude us, it is clear that the basic rationale underlying the choice of imagery is first of all a religious one, keyed, as I shall now discuss, on the faculty of the king to guarantee through the performance of rituals the benevolent presence of the local gods, and their teriomorphic manifestations.

Exemplary for this period is the image of the king libating in front of a god, an iconography strongly tied to the Late Bronze Age Hittite culture that will later on, significantly, entirely disappear from the local visual idiom. At Malatya, the 'potent king' PUGNUS-*mili* (II)45 materializes it in traditional Hittite fashion, reiterating on eight small-size blocks the image of himself, perhaps of his late grandfather46 and, once, of his wife47, all represented pouring libations in front of nine differ-

<sup>43</sup> These are the three relief orthostats filed by W. Orthmann under the style 'Malatya III' (Orthmann 1971: 94-95, Malatya C/1-3), to which we may now add two further reliefs found in situ at the «Porta Schaeffer» (Manuelli 2012: fig. 4). F. Manuelli (personal communication) has pointed my attention to the fact that that the corner orthostat Malatya C/1 has been later reworked and transformed into a stele, adding the image of a storm-god to an originally aniconic face (Malatya C/1b). This fact has not been recognized by W. Orthmann, who has been thus misled into a general later dating of this set of reliefs, leaving unexplained a series of close stylistic connections with the PUGNUS-*mili* reliefs (Orthmann 1971:95). On the bases of the iconographic development at Malatya and Karkemiš that I discuss in this article, I propose to re-organize the chronological sequence of the reliefs at Malatya according the order Malatya I – Malatya III – Malatya II, all of them dating before the destruction of the 'Porta Schaeffer' at the end of the 11th/beginning of the 10th century BC. The reliefs of the 'Malatya III' group can be further split into two stylistic sub-groups: 1. the heraldic reliefs with the armed *Mischwesen* (with stylistic parallel with a number of antithetical bull-men found at Karkemiš: see above, n. 63); and 2. the reliefs of the winged geniuses with pine-cone and a drooping plant on one side (with stylistic parallels in the relief Karkemis K/24 in Orthmann 1971). Since the first subgroup seems to have stricter stylistic affinities with the Malatya I group, it may be tentatively proposed to date the first subgroup earlier than the second subgroup.

<sup>44</sup> Passing animals, including bulls, sphinxes and winged lions, belongs to the earliest figurative decoration of the Water Gate at Karkemiš. At Malatya, winged geniuses, a passing bull, and armed *Mischwesen* were added to the PUGNUS-*mili* reliefs in the 11th century. Such animal arrays were part of the Late Bronze Age international language of the decorative arts, as a number of luxury items from Ugarit, Egypt, and elsewhere show (Feldman 2006: 25-58). Repeatedly combined with bull-men, however, they remind of specific arrangements – bordering on the notion of a cosmic diagram – developed in the Hittite milieu, and then adapted in the temple of 'Ain Dara: compare with the arrays of bulls, lions, fantastic animals and Atlantic figures at the spring sanctuary of Eflatun Pınar (Bachmann, Özenir 2004), on the Hittite ivory inlay from Megiddo (Loud 1939: no. 1; no. 44, pl. 11), on a ceremonial axe from Şarkışla (Bittel 1976a: 19-27), on a gold disc from Magnesia (Riemschneider 1954, Taf. 108), or on a stone plaque from Alaca Höyük (Bittel 1976b, fig. 246). A single relief block from Malatya, *Relief H*, is an odd one out, and represents a mythical battle between a Storm-God and a snake. The same iconography returns on a stele found at Tell Ashara, and may be tentatively interpreted as a symbolic transposition in myth of a discourse on supremacy and political struggle (Pecchioli 2001).

<sup>45</sup> While the kings of Karkemiš used the title MAGNUS.REX, other independent rulers of contemporary polities, without or with less direct links to the Hittite royal house, used as highest forms of address the title REX (Luw. \**hantawati-*; Giusfredi 2010: 82-88). At Malatya, PUGNUS-*mili* introduces the compound REX.\*462, perhaps to be translated as «Potent King» (Hawkins 2000: 307, following Meriggi).

<sup>46</sup> On *Relief E*, the caption «PUGNUS-*milii*» appears next to two different images of male figures: one of them is bearded and wears a horned, pointed hat; the other one is beardless, wears no hat, and sports long hair styled in peculiar curls. The PUGNUS-*mili* with beard and horned hat may be the dead grandfather of PUGNIS-*mili* (II), who sponsored the relief cycle (Brown 2008: 303). If this is the case, we shall assume that the royal title is assigned to the grandfather retroactively, since the first PUGNUS-*mili* appears to have been a REGIO.DOMINUS under the dominion of Karkemiš (as discussed above).

<sup>47</sup> On *Relief I,* a woman libates in front of the goddess Sauska. A Hieroglyphic Luwian caption identifies her as «Princess Tuwatis» or, alternatively, the «daughter of King Tuwatis» (Hawkins 2000: 287, n.72, referring to a proposal by M. van Loon). I favour the second interpretation, because Tuwatis seems to be a male name, and because the iconography is fit for a queen. What kind of reason would have PUGNUS-*mili* to represent, if any of his descendants, his daughter? But it makes sense to stress with the caption an inter-dynastic marriage with the daughter of a (Tabalian?) king whose name, we may presume, was conductive of political weight and influence.

ent gods48, each identified by his or her attributes and by Hieroglyphic Luwian captions. Evidently, the depiction of ritual practices linking the ruling dynasty with the community of the gods had a programmatic value. The steles from İspekçür and Darende and the Taita relief at Aleppo show that comparable images were also set up in other contexts, specifically in temples. In most cases, they were integrated by inscriptions indicating that these were commemorative monuments dedicated by the king to the gods upon the accomplishment of political deeds49. Although dedicatory inscriptions are not attested for gate reliefs, we may surmise that the occasion of their set up may have been similar. In the case of Malatya, the reliefs may relate to the early secession operated by PUGNUS*mili* (II) from the Land of Karkemiš and with his choice to enforce the newly acquired royal power exhibiting his cultic ties to the gods of the land. At Karkemiš, two large reliefs representing the king libating to the Storm-God and sitting at a ceremonial banquet were erected at the Water Gate – after a major destruction. If, as argued above, the libation and banquet diptych date more or less to the age of Taita, the devastation of the Water Gate and its earlier reliefs may be a material trace of Taita's presumed raid, and the libation and banquet relief part of a new gate decoration program (including basalt lions in 'Ain Dara fashion), and a celebration of the reinstated authority of the Great King.

The iconography of the worshipping king is emblematic of a public discourse that puts a great accent on the king, and on him alone, as the chief cultic performer. The theological background replicates ideas of the 13th-century Hittite Empire (cf. van den Hout 1995). Gods are represented in anthropomorphic form only face-to-face with the king (or, seldom, with a member of his family), almost as if only the king could truly summon and propitiate them. In these images, king and gods engage in a sort of private communication, whom the viewer contemplates as through a window. Much attention is paid to ritual gestures, hierarchy of appearances, cultic implements, symbolic attributes, and offerings, but virtually none to other individuals, who, if at all, are represented subordinated and instrumental to ritual procedures.

# *3. The balance of power in the art of the early 10th century BC*

At the beginning of the 10th century, monumental inscriptions at Karkemiš begin to illuminate the coexistence of the Great Kings with a line of local Country Lords, evidently reflecting the growing political influence of latter over the first. The Country Lords progressively imposed themselves as ruling subjects, and ultimately, in the second half of the 10th century, expedited the disappearance of the Great Kings from the political scene.

Earliest epigraphic records of this process are two analogous basalt steles of the early 10th century, set up by the Country Lord Suhi and his son Arnu-*x* in honour of the Great King Ura-Tarḫunta:


The steles commemorate in similar terms the military victory of Ura-Tarḫunta over the Land of Sura (Assyria?)50. We infer from their content that, at the beginning of the 10th century, Karkemiš was still governed by a Great King, who was also the leader of the army: «a dispute arose for him with the Land of Sura, and he opposed the army» (Karkamiš A4b, §2-3, Hawkins 2000: 80-81; Dinçol *et al.* 2014: 148). A Country Lord, specifically Suhi (I), operated next to him and bore the additional title of *tarwanis*, «ruler»51.

I follow F. Giusfredi in attributing a third stele found near Kelekli, 10km north of Karkemiš, to the same Suhi (Giusfredi 2014)52. The stele reproduces the iconography of the worshipping ruler typi-

<sup>48</sup> Cf. the «nine gods» (the Malatyan pantheon?) mentioned in Izgin 1, the inscription of CRUS+*RA/I* , perhaps to be read Tara, who was king of Malatya in the early 11th century (Hawkins 2000: 316, §18).

<sup>49</sup> Cf. Darende, §2-3: «I [Arnuwanti of Malatya] settled the city (-)tumani …, and I dedicated this stele to the god» (Hawkins 2000: 305).

<sup>50</sup> Weeden 2013: 10 *contra*, Simon 2012, proposing an identification with a region in Cappadocia.

<sup>51</sup> IUDEX-*ni*: a title attested only in the Iron Age, and specifically for the greater region around Karkemiš (Giusfredi 2010, 96). For an overview of the discussion, see Giusfredi 2009.

<sup>52</sup> In his recent important contribution, F. Giusfredi makes a convincing case for the existence of a Country Lord Astuwalamanza, father of Suhi I. *contra*, J. D. Hawkins, who prefers to attribute the stele of Kelekli to Suhi II, but admits that this creates «a historical puzzle» (Hawkins 2000: 93).

cal of the earlier centuries (it is in fact its latest attestation), only this time the ruler represented is not the King of Karkemiš, but the city's Country Lord53. In the first-person, Suhi lists his titles (REGIO. DOMINUS and *tarwanis*), states the name of his father (Astuwalamanza)54, and formally declares his intent to marry his daughter to an otherwise unattested «King Tutḫaliya», whom, we must presume, was Ura-Tarḫunta's son (Keleklİ, Hawkins 2000: 92-93; Giusfredi 2014: 490-491).

The steles of Suhi (I) and his son Arnu-*x* bespeak the growing aspirations and rise to power of the Country Lords. While still recognizing the authority of the Great King, Suhi takes possession of the royal idiom. He represents himself in ruling attitude and pursues attentive politics concerning his offsprings: the coronation of this politics must have been the marriage of his daughter into the legitimate royal family55, but he also foresaw the instalment of his son Arnu-*x* as priest of Kubaba and of his son Astuwalamanza (II) as his successor. Astuwalamanza (II) features in a fragmentary inscription on a portal lion as the ruler who presided over the building of a gate, presumably the great gate that led up to the citadel (Karkamiš A14b, Hawkins 2000, 83-87). Because of this inscription, F. Giusfredi (2014: 492) sees in Astuwalamanza (II) the first *tarwanis* of Karkemiš who acts as a ruler, in full independence from a Great King. Astuwalamanza's portal lion, however, was originally conceived to be located at the left hand of a gate jamb, and it comes in pair with a second, specular portal lion that bears an inscription of Astuwalamanza's son, Suhi (II) (Karkamiš A14a; Hawkins 2000: 83-87). Both portal lions share stylistic and palaeographic traits, and I am inclined to date both to a single commission by Suhi (II), who, in his inscription proclames:

«they [the gods] gave me authority […] my father and my grandfathers […] for them the gods did not exalt (their) person(s), but they exalted me» (Karkamiš A14a, §3-7, translation by J. D. Hawkins)56.

If this is correct, then it is Suhi (II) the first independent *tarwanis* of Karkemiš, his father Astuwalamanza (II) would have been the last Country Lord to operate under a Great King (Tutḫaliya?), and the political 'turn-over' shall be dated not earlier than the mid-10th century.

The proposed reconstruction leaves us with a period of 'political pre-redefinition' taking place in the first half of the 10th century and lasting about two generations, corresponding to the father-son dynasties of the Great Kings Ura-Tarḫunta and Tutḫaliya on one side and the Country Lords Suhi (I) and Astuwalamanza (II) on the other side. A conspicuous number of reliefs indicate that, in this period, public gates continued to be an important projection surface for visual messages. However, the Hittite iconography of the previous centuries was almost entirely abandoned, and the visual palette shifted to a whole range of new images. At Karkemiš, the *corpus* of reliefs dating to the first half of the 10th century consists of slabs found re-employed at the King's Gate and at the Herald's Wall – both built in the latter part of the same century (Gilibert 2011a: 38-49, with further references)57. Once again, the original set-up is unknown; at least some reliefs, and perhaps all of them, seem to have been conceived for an earlier version of the King's Gate.

The dominant tone of the images is still imbued with symbolism and religious connotations. Now, however, the accent is not any more on the royal state cult of Hittite tradition, but on the display of heroic force in combat and the hunt. The iconography of the worshipping king disappears altogether, while images of gods, demigods, demons, fighters and hunters in action multiply and intersect in heraldic forms. Emblematic for this phase, and quite specific to Karkemiš, are three-figure compositions on a single slab on which two mythical beings, or heroes, kill a fantastic animal, or an enemy, in a hand-to-hand

<sup>53</sup> Kelekli, §1: [… *kar-ka-*]*mi*? -*si-sa*(URBS) REGIO.DOMINUS-*ia-ix -sa*. The toponym *Karkamissa* is here determined with URBS, as opposed to REGIO, which is the archaic form. J. D. Hawkins has proposed to relate the habit of determining *Karkamissa* with URBS with the territorial loss of the late 10th century BC (Hawkins 1995: 90, n.32; Hawkins 2000: 74; Payne 2014: 151), which would be an argument for attributing the stele of Kelekli to Suhi II. However, in the 10th century BC the use of the determinative URBS instead of REGIO and *vice versa* seems to have been less than univocal: for example, Katuwa, the son of Suhi (II), employs URBS as a rule, but at least in one case declares himself «Lord of *Karkamissa*(REGIO)»: Karkamiš A2, §1.

<sup>54</sup> Formerly spelled Astuwatamanza: for the new reading, see Rieken and Yakubovich 2010: 203. This is the «Astuwatamanzas Zero» identified by Giusfredi 2014.

<sup>55</sup> This marriage, if ever took place, may be at the roots of the fact that, beginning with Suhi (I) down to Katuwa (end of 10th century) Country Lords presented themselves as 'kinsmen' of the Great Kings (for Suhi I, see the new stela in Dinçol *et al.* 2014: 150; for Katuwa, cf. Karkamiš A11b-c and A11a).

<sup>56</sup> This *topos*, of which the quoted passage is the earliest attestation, seems to belong to a new royal idiom. It returns, in modified forms, in inscriptions of Suhi's son, Katuwa (Karkamiš A2, Karkamiš A12, Karkamiš A25), on the statue of Halparuntiya (II) of Gurgum (Maraş 4, mid-9th century), as well as in the inscriptions at Tell Halaf (Kapara, c. 900) and Zincirli (second half of 9th century).

<sup>57</sup> At Malatya, these set of reliefs has its contemporary counterpart in the reliefs of the 'Malatya II' group, consisting of three slabs with hunting scenes, integrated by Hieroglyphic Luwian inscriptions (Orthmann 1971: 93).

combat58. The interest for three-figure contest scenes is best observed on four large square slabs, each an elegant variation of the same theme (Fig. 5). These images, as well as most reliefs of this period, have nothing 'Hittite' about them, but rather go back and elaborate on earlier North-Syrian patterns. Their antecedents are best sought in the hunt and contest scenes on Mittanian and Middle Assyrian seals59. One of the seals of Ini-Teššub (Fig. 6), viceroy of Karkemiš under Tutḫaliya IV, shows that, at Karkemiš in the late 13th century, such contest scenes were already part of the local royal self-representation, and were employed as a complement to the image of the worshipping king (Mazzoni 1977: 13-14; Beyer 2001: 48-49). The artists of the early 10th century call upon this visual tradition, which evidently lived a parallel, and partially hybrid, life next to the Hittite idiom. Its monumental reformulation in the Iron Age introduces a significant interest for tripartite contests where the main contestants are not fighting over their prey, but collaborating to kill it – a variant of the contest scene which enjoyed only a limited favour in the later 2nd millennium BC (Matthews 1990: 104-105). How shall we understand the switch to this different visual palette in the decoration of urban spaces? As we have seen, the Great Kings started using gates for the public representation of the negotiation of royal power and divine grace; in doing so, they used signature images of the Hittite Empire. I propose to relate the abandonment of these signature images and the change of visual idiom to the introduction of a new discourse in public propaganda. Although the mythological narratives behind the triadic combats elude us, it is quite safe to assume that these images were charged with allegorical meanings. The basic theme reiterated over and over again is the balance of power between two parties, who fight side-by-side to overcome a common enemy. Perhaps we may put it like this: at a time when military conflicts were steadily increasing, and territorial control progressively contracted, the imperial pretensions of the Great Kings of Karkemiš lost ideological grip, and grew increasingly untenable. At the same time, epigraphic sources indicate that, for reasons that are still to be explored, the Great Kings placed more and more power in the hands of their governors, the Country-Lords. Taking into consideration this political background, we may begin to understand why, in public contexts, the last Great Kings of Karkemiš resorted to images of propaganda keyed on cooperation and control, with a world 'out there' wild and untamed.

#### *4. Conclusions*

The reliefs at the King's Gate and at the Herald's Wall are the last examples of the art of the Great Kings. Next comes the art of the independent Country Lords, who *de facto* invent a novel political identity, change ritual practices and redefine criteria of visual propaganda60. In the second half of the 10th century, Suhi (II) and his son Katuwa relocate public art from the urban gates to the walls lining the central square of Karkemiš, transforming it into an arena for mass ceremonies and initiating great figurative cycles, dominated by images of civic ceremonies and military triumphs. Warriors, women, young people are represented together with the king and the queen in joint celebration. The reliefs are accompanied by longer Hieroglyphic Luwian inscriptions reporting the civic and military deeds of the ruler, and including prescriptions for offerings to his image: statues of rulers are erected nearby, with installations for the performance of offering rituals. As S. Aro (2013: 236-244) and S. Mazzoni (2013: 475-476) recently spotlighted, the new monumental and epigraphic habit also forsaw the 'capitalization' of the incipit of inscriptions, the Hieroglyphic Luwian sign EGO (*amu*-, «I am»), into the full-blown figure of the ruler in the gesture of speaking. In short, the public art of the Country Lord initiates a new cult of the ruling dynasty, with markedly mundane images appealing to a sense of belonging and addressing the audience directly, in fact actually *speaking* to the audience, and requiring regular offerings. We may relate the remarkable effort invested by Suhi and Katuwa into building this elaborate machinery of propaganda to their need to cement public consensus in and around the city, where conflict was rampant.

<sup>58</sup> Related iconographies are also found at the Palace of Kapara at Tell Halaf and, of more difficult interpretation and much later in date, at the gates of Karatepe.

<sup>59</sup> The best discussion of these antecedents is Mazzoni 1977: 15-20; N.B. the important reference to the 14thcentury seal of Aššur-mutakkil (Mazzoni 1977: 19).

<sup>60</sup> Once again it is interesting to compare with the situation at Malatya, where the hunting scenes of the early 10th century (the 'Malatya II' group) are the last traces left by an iconographic tradition tightly tied to the Hittite world of the Late Bronze Age. The destruction of the gate where they were originally set up was followed by a gap period of at least two generations, during which the ruins were occupied by squatters. A century later, the fortifications were re-built and the architects took care to integrate in them *spolia* of the older gate decoration, but all significant ties to Hittite material culture had forever been severed (Manuelli 2014).

#### 148 *ALESSANDRA GILIBERT*

A long and prominently exhibited inscription of Suhi, Karkamiš A1a, records Suhi's military triumph over an otherwise unknown Hatanima, but also reveals that Hatanima was able to raid the city and ravage its temples (Hawkins 2000, 87-89). Even more indicative of political stress are two inscriptions of Katuwa, Karkamiš A11a (Karkamiš A11a, §5-6) and A11b+c (Hawkins 2000: 94-108), which imply that the «grandsons of Ura-Tarḫunta» staked claims upon the throne of Karkemiš until the end of the 10th century, when a certain Ninuwi organized a revolt, gained the military support of the hinterland, and took possession of the city. Afterward, so claims Katuwa, «the city lie desolate» (Karkamiš A11b, §3)61: according to the new reading proposed by C. H. Melchert (2011: 77), Katuwa in some way acknowledged the claims of Ninuwi, and actually bought him out of the city by means of some sort of legal transaction62. And this is the last we hear of the progeny of the Great Kings of Karkemiš.

#### *Bibliography*

Abū 'Assāf A. 1990, *Der Tempel von 'Ain* Dārā, Mainz.

Alexander R.L. 1989, *A Great Queen on the sphinx piers at Alaca Höyük*, «Anatolian Studies», 39: 151-158. Archi A., Venturi F. 2012, *Hittites at Tell Afis*, «Orientalia», 81: 1-55.


Bunnens G. 2006, *A new Luwian stele and the cult of the Storm-God at Til Barsip-Masuwari*, Louvain.


Del Monte G.F. 1973, *La porta nei rituali di Boğazköy*, «Oriens Antiquus», 12: 107-129.

De Martino S. 2010, *Symbols of power in the Late Hittite kingdom*, in Y. Cohen, A. Gilan, J.L. Miller (eds.), *Pax hethitica: studies on the Hittites and their neighbours in honour of Itamar Singer*, Wiesbaden: 87-98.

<sup>61</sup> For the translation, I follow Melchert 2011: 76.

<sup>62</sup> C.H. Melchert (2011: 77) ingeniously suggests that Katuwa traded Karkemiš for other cities. I venture to suggest that that was the beginning of one of the numerous territorial secessions that dismembered the Land of Karkemiš in the course of the 10th century. The resolution of political conflict by diplomatic and legal means is in line with the monuments erected by Katuwa to commemorate this accomplishment: Katuwa's figurative cycles are pervaded by a sense of civic harmony, and avoid images of fight and violence (as opposed to the macabre parades of dead and dying enemies of his father Suhi). It is also significant that the independent Country Lords did not remove the monuments of the Great Kings from the city's public spaces – indeed, as in the case of the Herald's Wall, they even gave them new importance. It is also telling that the stele of Arnu-*x*, a written testimony to the original loyalty of the Country Lords to the Great Kings, was left on exhibit in the courtyard of the Storm-God temple throughout the city's history. The same may be said of the survival of the Late Bronze Age Hittite royal jewels that have been found in the Iron Age Gold Tomb of Karkemiš (probably 7th-6th century BC; Woolley 1952: 250) – apparently the jewels had been treasured over centuries, probably not only for their material value, but also for their symbolic significance. In short, the extant evidence coherently indicates that the Country Lords of the second half of the 10th century integrated the previous regency of the Great Kings in their claim to a legitimate rulership.


#### 150 *ALESSANDRA GILIBERT*

*in vassal treaties from the 13th Century B.C.E*, in C. Wilcke (ed.), *Das geistige Erfassen der Welt im Alten Orient – Sprache, Religion, Kultur und Gesellschaft*, Wiesbaden: 191-207.


Woolley C.L. 1921, *Carchemish. Report on the excavations at Jerablus on behalf of the British Museum. Part 2: the town defences*, London.

Woolley C.L. 1952, *Carchemish. Report on the excavations at Jerablus on behalf of the British Museum. Part 3: the excavations in the inner town*, London.

Woolley C.L., Barnett R.D. 1952, *Carchemish. Report on the excavations at Jerablus on behalf of the British Museum. Part 3: the excavations in the Inner Town and the Hittite inscriptions*, London.

Yener K.A., Dinçol B., Peker H. 2014, *Prince Tuthaliya and Princess Ašnuhepa*, «N.A.B.U*.*», 4: 136-138.

Fig. 1. Great Kings and Country Lords at Karkemiš between 1200 and 900 BC: a provisional overview following Giusfredi 2014.

Fig. 2. 11th-century portal lion from 'Ain Dara (left) and fragment of a nearly identical portal lion found at the Water Gate of Karkemiš (Wolley 1952, fig. 32).

Fig. 3. *Relief K* from the *Porte des Lions* at Malatya, 12th century BC (Hawkins 2000, Malatya 8) and the 'libation slab' from the Water Gate at Karkemiš, 11th century BC (Wolley 1952, Pl. B. 30a).

Fig. 4. Winged lion from the Water Gate and coeval sphinx found in secondary context at Karkemiš (Woolley 1952, Pls. B. 29b, B. 48a)

Fig. 5. Three-figure contest scenes from the Herald's Wall (Wolley 1952, Pls. B. 11a, B. 15a-b, B. 16a).

Fig. 6. Seal of Ini-Teššub, king of Karkemiš under the Hittite Great King Tutḫaliya IV, late-13th century BC (Schaeffer 1956, fig. 32).

# CONTRIBUTORS

Alfonso Archi Via Montevideo 2 A 00198 Roma (Italy) alfonso.archi@gmail.com

Francesco G. Barsacchi University of Florence Department of History, Archaeology, Geography, Fine and Performing Arts (SAGAS) Via San Gallo 10 50129 Firenze (Italy) francescobarsacchi@hotmail.com

Massimo Forlanini Via Claudio Monteverdi 5 20131 Milano (Italy) mforlanini@hotmail.it

Niccolò Galmarini via Giannozzo Manetti 15 00167 Roma (Italy) nicogalma@virgilio.it

Alessandra Gilibert Università Ca' Foscari Venezia Dipartimento di Studi Umanistici Dorsoduro 3484/D, 30123 Venezia (Italy) Alessandra.Gilibert@topoi.org

David Hawkins 18a Ridgmount Gardens London WCIE 7AR (United Kingdom)

Andreas Müller-Karpe Vorgeschichtliches Seminar der Philipps-Universität Marburg Biegenstr. 11, 35037 Marburg (Germany) muekarpe@staff.uni-marburg.de

A. Tuba Ökse Kocaeli University Kocaeli Üniversitesi Fen-/Edebiyat Fakültesi B Blok, Arkeoloji Bölümü, Umuttepe Yerleskesi, TR-41380 İzmit/Kocaeli (Turkey) tubaokse@yahoo.com

#### 158 *SACRED LANDSCAPES OF HITTITES AND LUWIANS*

Andreas Schachner Deutsches Archäologisches Institut, Abteilung Istanbul Inönü Cad. 10, 34437 Istanbul (Turkey) andreas.schachner@dainst.de

Aygül Süel Ankara Üniversitesi, Dil ve Tarih-Coğrafya Fakültesi, Hititoloji Anabilim Dalı Department of Ancient Languages and Cultures - Hittitology 6100 Sıhhıye/Ankara (Turkey) aygulmustafasuel@gmail.com

Mustafa Süel Uludağ University, Tarih Bölümü Department of Ancient History Salkım Söğüt sok. 15/9 06460 Dikmen/Ankara (Turkey) aygulmustafasuel@gmail.com

Piotr Taracha University of Warsaw Faculty of Oriental Studies - Department of Ancient Near Eastern Studies Krakowskie Przedmieście 26-28, 00927 Warszawa (Poland) piotr.taracha@uw.edu.pl

Gernot Wilhelm Julius-Maximilians-Universität Würzburg Mozartstr. 2a 97209 Veitshoechheim (Germany) grnt.wilhelm@t-online.de

# STUDIA ASIANA

# *A series established in 2001 by Alfonso Archi, Onofrio Carruba and Franca Pecchioli Daddi.*

