# Rulership and the Gods: The Role of Cultic Institutions in the Late Bronze to Iron Age Transition in Anatolia and Northern Syria1

Lorenzo d'Alfonso, Nathan Lovejoy

**Abstract**: This paper aims to demonstrate that cults and cultic institutions are a crucial element for understanding the processes producing different regional outcomes after the fall of the Hittite empire. In this paper, cults are understood as normative cosmic forces defining tempo and worldview of ancient societies. Cultic institutions can be identified as physical spaces defined by purity, charged with real and symbolic value, and led by specialists whose competence is recognised by the community. Instead of being a by-product of political complexity, they are a driving force behind the power dynamics because they are perceived as such in a bottomup perspective, but also often by main political actors in search of legitimation of their power. This paper examines the interconnections between cultic and political institutions in the territory under the Hittite empire and in the same space after the empire's demise. We aim to distinguish between processes of resilience, reorganisation, and transformation as they occurred in particular micro-regions previously controlled by the empire, including the Upper Euphrates, South-Central Anatolia, North-Central Anatolia, Cilicia, and the Northern Levant; this will demonstrate both the importance of such a micro-regionally defined study, as well as the shared coincidence of cultic and political institutional change. It will become evident that cultic continuity coincided with the resilience of political institutions, and changes in the cultic landscape corresponded to political reorganisations or transformations in post-Hittite Anatolia and north Syria.

1. Question of the source and nature of political control and political authority in territories of the former Hittite empire

In recent years, archaeology has contributed tremendously to a lively one hundred years long debate among text-based historians of ancient western Asia and Indo-European linguists on the sources and the nature of political control and of political authority in the Hittite empire (for the contribution of archaeology see e.g. Glatz 2009; 2020; Schachner 2011; 2009; d'Alfonso 2021; for text-based contributions see Beckman 1995; Cancik 1993; Gilan 2004; Archi 1979; Goetze 1957; 1964; Gurney 1980; Imparati 1988; Klengel 1988; 1999; Otten 1964; Starke 1996; Sürenhagen 2001; lastly de Martino 2022, with literature therein). The fragility characterising ancient polities (but the formula 'ancient polities' discounts the reality of modern polities we are living in) is even more evident in the case of the Hittite empire than any other western Asian polity. Textual sources as well as archaeological, geo-archaeological, and bio-archaeological evidence indicate the existence of geo-morphologically and environmentally challenging settings within the Anatolian Plateau, the high level of diversity in habitus between neighbouring communities living in the Plateau un-

<sup>1</sup> This work is the result of a continuous collaboration, however, LDA is responsible for parts 1, 2.2, and 2.3, while NL is responsible for parts 2.1, 2.4, and 2.5; the conclusions were drafted together.

Lorenzo d'Alfonso, University of Pavia, Italy, lda5@nyu.edu, 0000-0001-9163-3417 Nathan Lovejoy, New York University, United States, ncl291@nyu.edu, 0000-0002-0124-5398

Referee List (DOI 10.36253/fup\_referee\_list) FUP Best Practice in Scholarly Publishing (DOI 10.36253/fup\_best\_practice)

Lorenzo d'Alfonso, Nathan Lovejoy, *Rulership and the Gods: the Role of Cultic Institutions in the Late Bronze to Iron Age Transition in Anatolia and Northern Syria*, © Author(s), CC BY 4.0, DOI 10.36253/979-12-215-0042-4.11, in Clelia Mora, Giulia Torri (edited by), *Administrative Practices and Political Control in Anatolian and Syro-Anatolian Polities in the 2nd and 1st Millennium BCE*, pp. 177-214, 2023, published by Firenze University Press, ISBN 979-12- 215-0042-4, DOI 10.36253/979-12-215-0042-4

der the Hittite *aegis*, and the endemic conflicts characterising the access to political power (lastly de Martino, Devecchi 2020; Schachner 2022). This challenging, divisive element notwithstanding, the royal dynasty of Hattusa succeeded in reaching and maintaining political authority over the entire Anatolian Plateau for *c.* 400 years. Archaeological evidence shows that the pillar which sustained the Hittite Great Power is the invention and construction of imposing infrastructure for the storage and usage of water and grains (Schachner 2011). Two other kinds of infrastructure characterise the Hittite empire, namely temples and fortifications (Zimmer-Vorhaus 2011; Mielke 2011; 2018). Remains of the four types of infrastructure have been excavated at most urban sites in the core region of the empire, north-central Anatolia, but also beyond it. While some water catchment infrastructures were also built in the countryside, granaries and other water storage facilities were erected within the perimeter of the fortification walls in urban centres. In a way, this infrastructure can be seen as a further means for storing essential products, not against conditions caused by natural instability, but against conditions produced by human instability, such as robbery and conflicts. A similar function could be attributed to temples. We are aware of cultic space in extra-urban contexts (Ökse 2011), but with few exceptions, most temples were located within the fortification walls perimeter in urban context. In an indigenous perspective, as seats of god's *simulacra* and house to the gods themselves, temples were instrumental for obtaining the gods' intervention regularly requested for protecting primary production and military activity, thus reinforcing the core pillars of the political power in the protection of primary means of life: grain and water (d'Alfonso 2021).

In this section we aim at summarising some relevant features on the cult during the Hittite empire. From early on, the study of the cuneiform texts had underscored an investment in establishing laws and administrative procedures. The normative effort is particularly evident in the cult, with ritual texts registering materials and procedures to be adopted in an array of specific circumstances, festival organization texts providing information on locations and schedules of cultic celebrations, and cultic inventories supporting the economic and administrative functioning of each cultic event (Cammarosano 2018). The relation between the cult and political authority is ubiquitous. In the Hittite empire one cannot be separated from the other. As to the Hittite royal dynasty, its legitimation to monarchic power was based on divine sanctioning. Equally, the territorial spread of this sacred kingship was reinforced on the one hand by the centripetal gathering of local and foreign gods, their *simulacra*, and their liturgic texts in the archive of the royal family in the Hittite capital, and on the other hand by the 'centrifugal' methodical organization of cultic journeys of the royal family to a number of cultic institutions outside of the Hittite capital, in north-central Anatolia and in other regions of the Plateau. The celebration of festivals at Hattusa and elsewhere in the Hittite territory emerges from the cuneiform texts as the main activity defining time and seasonality and political and collective activity in Hittite Anatolia. Besides their spiritual investment, it required an investment of resources that is not possible to quantify for the lack of economic texts, but that was far more expensive than any other activity that we are aware of except possibly for war (lastly Cammarosano 2018; d'Alfonso, Matessi 2021; Schwemer 2022: 387-92).

Cultic institutions such as temples and sanctuaries were essential in this process defining political authority in Hittite Anatolia, both in the case of festivals celebrating the gods at the capital Hattusa and those celebrating the gods in other urban centres, or non-urban sanctuaries (Archi 2015). Archaeologically, a number of these cultic institutions whose festivals are also mentioned in the texts have been excavated. A large number of temples have been exposed at the Hittite capital Hattusa; here, a topography of the cults has been attempted, but identifications are still very open (Zimmer-Vorhaus 2011; Schachner 2011; Wilhelm 2011). The same holds true for the *sancta sanctorum*

of Yazılıkaya, whose identification with a specific monument still escapes us (Seeher 2011). More promising are the identifications of temples and sanctuaries in other urban and extra-urban centres. The temple of the Storm-god of Nerik and the sacred spring mentioned in the cultic sources have been recognised as the temple and the megalithic, underground stairway excavated at Oymağaç Höyük (Ökse, Czichon, Yılmaz 2021: 241-243, with literature therein; Czichon *et al.* 2011). At Kuşaklı, Sarissa, like in Hattusa the urban topography of the cult is still vague, while a sanctuary of the Storm-god of the *ḫuwaši*- celebrated in a festival, has been associated with the poor architectural remains identified next to a pond 2.5 km away from the town in the countryside (Wilhelm 2015). The megalithic foundations of the monumental building on the northern terrace of Uşaklı Höyük has been identified with a Hittite temple, which seems secure, and its identification with the temple of the Storm-god at Zippalanda is likely (D'Agostino, Orsi, Torri 2019-2020, with previous literature). Even though characterised by a certain level of variation, the organisation of the space in the cultic buildings within the core of the Hittite empire presents characteristic features, such as the presence of a central rectangular court with one or two porticoed consecutive sides, a *cella* protruding out of the perimeter of the building, often in a prospicient position, at the opposite side of the gate or the gates giving access to the building, and one or more rows of small rooms running all around the central court and used as storerooms for different sorts of movable goods (Fig. 1). The spatial organisation implies the urban prominence of the building, with the external visibility of its most sacred space: the *cella*. Characteristic is also the presence of a ceremonial spaces internal to the building hosting a selected, but still representative, urban public for rituals and festivals. Functional analysis implies that temples also worked as collective repositories for ceremonial repast and other economic activities (Zimmer-Vorhaus 2011; Herbordt, von Wickede, Schachner 2021).

Despite limited archaeological evidence, there is no doubt regarding the centrality of the temple institutions all over the central Anatolian Plateau in the organisation of central power and local communities during the Hittite empire. The evidence for the celebration of festivals throughout the Plateau by the royal family in association with cultic personnel and representatives of local communities covers the whole history of the Hittite empire (Schwemer 2022, with previous literature). Similarly, the plans of the temples, while again subject to variation over time, maintain their basic elements and the two aspects of celebrative gathering and storage along the whole Hittite period. This is even more astonishing if one considers that the administration of the Hittite Great Power changed consistently through time, both in scale and in defining criteria (Matessi 2016). This once again may reinforce the idea of the cultic institutions as a steady actor in defining and negotiating political authority in Hittite Anatolia.

Starting with the late 14th century, the Hittite empire was able to win Cilicia and Syria under its hegemonic control. This vast region south of the Taurus was at the time fragmented in a number of small polities and the form of political control over them varied depending on the modalities of the positioning of each of them at the time of the Hittite Syrian campaigns, but also on the balance of power and the distance or proximity from other Great Powers. As a matter of fact lands such as Kizzuwadna, Aleppo, and Karkemiš were secundogeniture assigned to royal princes of the Hittite dynasty from Hattusa, and also in the land of Mukiš the local rulers were substituted by a member of the Hittite royal family (Klengel 1992, part III.2; 2001; d'Alfonso 2011; lastly Devecchi 2022: 277-278). At least for the first 40 years after the campaign, Aleppo and Kizzuwadna, the two most prominent religious centres south of the Taurus, were led by one and the same dynasty. Even if the new dynasty in Aleppo enjoyed the title of King (LUGAL), cuneiform and hieroglyphic sources show that they were more often

named with the title of priest, placing the cultic element of their authority in the foreground (d'Alfonso 2007). Around 1300 BCE, a similar situation seems to characterise the land of Aštata. Here, cultic paraphernalia with Hittite imagery and cultic inventories hinting at the celebration of Hittite gods in town implies the establishment of a direct contact of the core of the empire with this remote urban centre on the Middle Euphrates through the organisation of the cult. The diviner of Emar became the most relevant representative of Hittite interests in town, first directly for the Great Kings of Hattusa, and from the mid-13th century for the Syrian secundogeniture of Karkemiš (Fleming 2000; Cohen, d'Alfonso 2008; Cohen 2010; Archi 2014; Thames 2020; Strosahl 2022).

Now looking at the temples, it is worth noting that neither in the centres in which local rulers kept control of political authority, nor in centres in which members of the Hittite royal family had been installed as local rulers, the organisation of the architectural space resembles the one of the temples in Hittite Anatolia. The planimetry is markedly different and based on one of two local traditions of *in antis* temples, without an inner open court and rows of storerooms, but consistently associated with a court, a terrace, or any other type of open space for cultic performance (Wightman 2007: 22, 159-173; Otto 2013; Mazzoni 2015; Stroshal 2022: 233-235) (Fig. 2).

Thus, if the criteria defining the architectural space organisation of the central Anatolian Hittite temples are lacking in Hittite controlled centres south of the Taurus,2 some distinct Hittite elements exist in certain features. We refer to the monumental stone art in the form of figurative reliefs and/or inscriptions placed at or next to the main temples of the centres under the authority of new Hittite dynasties. At Sirkeli Höyük, identified with Kummanni/Lawazantiya, the rock relief of Great King Muwatalli, and possibly the one of Mursili III next to it, was carved on the rock outcrop in which the great temple building of the city has been identified (lastly Novák 2019-2020: 152-156, with reference

<sup>2</sup> With the exception of Tarsus.

therein). At Alalah, the orthostat with the figurative relief of the Prince and Priest Tuthaliya together with his wife has been found reused in a later phase of the temple in the citadel, but there is little doubt that the relief orthostat was originally placed at the gate providing access to the temple, which was reorganised after the Hittite takeover (Yener 2013). Of the LBA temple of the Storm-god in the citadel of Aleppo we know very little, but the dedicatory inscription in Anatolian Hieroglyphic by king Talmi-Šarruma for the construction of the temple of Hepat and Šarruma is proof of the adoption of the same monumental art associated with the Hittite power in the three sites. We might add that also at the temple of 'Ain-Dara there is evidence of Hittite inspired monumental reliefs, such as the LBA II figurative orthostat of Šauška, a slab with the lower portion of another figurative orthostat preserving the legs of a male figure and poor remains of a monumental Hieroglyphic inscription, and a relief fragment with a head interpreted as a Great King. These pieces support the existence of a LBA II construction phase of the temple before the two main IA phases extensively investigated by scholars, and the adoption of monumental art associated with the Hittite empire (Abu Assaf 1990; Kohlmeyer 2008; Novák 2012; Aro 2022: 563). Even at Karkemiš, N. Marchetti has recently suggested that the so-called Ḫilani temple, already existing during the LBA I before the Hittite conquest, was embellished with figurative orthostats during the LBA II, when the new Hittite dynasty was installed in the town. Of its program we know very little, as only two reliefs of bull-men facing in opposite directions have been assigned to this program, and the main figures are lacking (Marchetti 2019-2020: 271). While the use of figurative art and of hieroglyphic inscriptions is not unknown in the core of the Hittite world as the Tuthaliya stele from Temple 5 at Hattusa, of the reliefs from Örtaköy-Sapinuwa, and above all of

Fig. 2. Plan of temples at 'Ain Dara (from Novak 2012: 47 fig. 4 [drawing by G. Elsen-Novák after Abu Assaf 1990, Abb. 18]) and Emar (Margueron 1995: 132).

the rock sanctuary of Yazılıkaya, these are exceptions and most of the excavated Hittite temples and cultic building in the central Anatolian Plateau are not supported by a monumental iconographic and/or graphic apparatus on stone (Fig. 3).

By contrast, all temples belonging to major centres in which Hittite officials and a Hittite administration were installed as chief local authority kept the organisation of space and function preceding the Hittite takeover but added an iconographic apparatus on stone. In the cases of Alalah, Aleppo, 'Ain-Dara, and Sirkeli Höyük, together with a figurative component, there is also an epigraphic component written in Anatolian Hieroglyphic that serves almost exclusively to indicate the patronage of the new Hittite authority in town over the local cultic institution. On the Anatolian Plateau this is exclusively attested at the sanctuary of Yazılıkaya, but it is worth nothing that the figurative and epigraphic apparatus indicating the patronage by Great King Muwatalli (and Mursili III?) at Kummanni, Prince Tuthaliya at Alalah, and King and priest Talmi-Šarruma at Aleppo precede by some 50 years those of Great King Tuthaliya in Chamber A and B at Yazılıkaya, so that this may well have been a political strategy of emphasising patronage of a cultic institution originating in the political experience of the reorganisation of power in the Syrian and Levantine city states, rather than an Anatolian tradition. In Anatolia, similar patronage on stone is expressed on rock-carved, extra-urban landscape monuments, not in the urban cultic space. It is possible, even likely, that other supports, maybe perishable, and other strategies (such as offering for deceased, divinised kings) would make such patronage well present in the core of the empire, as well. This notwithstanding, monumental figurative and epigraphic compositions as decorations of urban buildings devoted to the main urban cults are a form of expression of the patronage of Hittite newcomers in the region south of the Taurus.

2. Processes of resilience, reorganization, and transformation in Iron Age cultic and political institutions

Back in 1983, in a paper devoted to the post-Hittite developments in Anatolia, E. Akurgal reported the reconstruction of the time inherent to the political situation after the fall of the Hittite empire:

Es scheint tatsachlich, dass Anatolien in der Zeitspanne zwischen 1180 und 750 v. Chr. von nomadischen Völkern besiedelt war, die keine festen Sitzplatze hatten. Die phrygischen Oberreste liegen zwar in Hattusa direkt auf der hethitischen Schicht des 12. Jh. Es ist jedoch kein Gegenstand ans Tageslicht getreten, der alter zu datieren wäre als um 750 v. Chr. Das ist eine wichtige Feststellung. Denn solange wir keine Kulturreste innerhalb des Halysbogens finden, die über das 8. Jahrhundert hinaufreichen, sind wir genötigt anzunehmen, dass dieses Gebiet 400 Jahre lang nur schwach besiedelt war oder überhaupt Besitz der Nomadenstamme wurde … Ob diese nur auf die kriegerischen Ereignisse der Volkerwanderung oder auch auf vernichtende Krankheiten wie die Pest, die im Elend der Nachkriegsjahre vielfach zu herrschen pflegte, oder auf Perioden der Trockenheit, die in diesem Gebiet nicht selten vorkommen, zurückzuführen ist, wissen wir nicht. Eines ist aber sicher, dass in dem Kerngebiet des Hethiterreiches bis zur Entstehung des Phrygischen Reiches um 750 v. Chr. Keine städtischen Siedlungen vorhanden waren, und mit großer Wahrscheinlichkeit hat es als Weideland der Nomadenstamme gedient. (Akurgal 1983: 75)

In 2004, as H. Genz wrote his introduction on the EIA in Central Anatolia, this view was still prevalent in the archaeological literature: 'Langer Zeit galt es als allgemeine akzeptierte Tatsache, dass die eisenzeitlichen Befunde in Zentralanatolien nicht vor dem 8.

Fig. 3. Hittite royal reliefs from the temple of Alalah (Tudhaliya and Ašnu-hepa; photo by N. Lovejoy) and Yazılıkaya (Tudhaliya and Šarruma; Bilgin) and Sirkeli (Muwattalli; Bilgin).

Jh. v. Chr. einsetzen' (For a long time it was generally agreed that there are no evidence of Iron Age finds in central Anatolia dating before the 8th century BCE) (Genz 2004: 36). Evidence from the Upper Euphrates and from northern Syria supported the existence of complex political entities whose formation started in the immediate period after the demise of the Hittite empire. The contrast between the lack of evidence for the territory corresponding to the former core of the empire, opposed to the evidence for Syria and the Upper Euphrates, was at the basis of the thesis of a mass migration from north-central Anatolia to the region south of the Taurus. This thesis, despite some criticisms and with useful revisions, is still maintained in current scholarship (Osborne 2021: chapter 2, with reference therein). G. Summers' recent review of the post-Hittite developments north of the Taurus has thus developed the long accepted view in the following ways: 1) political complexity in central Anatolia starts only in the mid-9th century BCE; 2) political complexity is a consequence of the expansionism of the Assyrian empire; and 3) it is a kind of secondary state formation inspired by the model of Assyrian imperialism and the Syro-Hittite political culture of the northern Levantine and western Syrian region (Summers 2017).

Summers' picture may be adequate for some areas, but a micro-regionally based reconstruction is more productive in revealing a variety of modalities and outcomes characterising the transition from the Hittite empire to new forms of political organisation in the post-Hittite period. We suggest that after the fall of the Hittite empire, its former territory fragmented into micro-regional outcomes characterised by different processes of political resilience, reorganisation, and transformation (d'Alfonso 2020a). In this view, the micro-regions of north-central Anatolia and west-central Anatolia are characterised by intense political transformation, the Upper Euphrates area is characterised by political resilience, whereas in south-central Anatolia the EIA outcome has been associated with a process of reorganisation. Cilicia and the northern Levant, although at the core of intense attention by archaeologists, have not been investigated under this perspective, and this will be done in the context of the present paper. We will concentrate our analysis on the evidence of non-private, cultic, sacred spaces after the transition in each micro-region, with the goal of examining the role of cultic institutions in the processes of resilience, reorganisation, and transformation.

# 2.1. Resilience in the Upper Euphrates

The Early Iron Age kingdoms of the Upper Euphrates region demonstrate substantial resilience following the fragmentation of the Hittite empire, as seen through the continuity of cultic and political institutions in Karkemiš and Malatya. Archaeologically, this continuity is evident at Karkemiš in the persistence of Temples A and B (the so-called Ḫilani Temple), apparently dedicated to the Storm-god and perhaps Nikarawa during the Iron Age, and possibly also before, and Temple AA on the northwest acropolis, likely dedicated to Kubaba (Woolley-Barnett 1952: 169-182; Marchetti 2014: 315-317; Marchetti, Peker 2018; Marchetti, Peker, Zaina in Marchetti (ed.) 2019-2020: 269-272; Hutter 2021: 212-215), as well as in the renovations to the Lower Palace and other monumental administrative structures at the site (Marchetti, Peker, Zaina in Marchetti (ed.) 2019-2020: 270-271, 274-276), including what appears to be an Early Iron Age palace recently discovered beneath the later 10th century BCE palace of Katuwa (Fig. 4).3 This

<sup>3</sup> Notice of the new discovery was given at a conference presentation: Marchetti N. 2022, Recent archaeological discoveries on the Late Bronze and Iron ages at Karkemiš, at the PALaC (Pre-Classical Anatolian Languages in Contact) workshop, *Languages and Cultures in Contact in the Ancient Mediterranean. Verona, March 24-25, 2022.*

Fig. 4 Plan of Karkemiš highlighting temples and adjacent monumental space (Marchetti et al. 2019-2020: 332 fig. 1.3; 359 fig. 8.3).

evidence is supported by epigraphic finds that demonstrate the continued use of Late Bronze Age titulary along with the persistence of the Hittite dynastic line at least into the 12th century BCE with Kuzi-Teššub, (Great) King and Hero of Karkemiš, identified on his seal impression (Hawkins 1988, with references therein; Hawkins, Weeden 2016: 9-10), as well as in multiple Hieroglyphic Luwian inscriptions from the Upper Euphrates region (GÜRÜN; KÖTÜKALE). Exogenous Assyrian sources also reflect a continuity in foreign perceptions of the space, which Tiglath-Pileser I defines as 'the land of Hatti' (RIMA 2, A.0.87.1: v 49; A.0.87.2: 28; D'Orazio *et al.* in Marchetti (ed.) 2019- 2020: 274). While properly Early Iron Age reliefs and inscriptions are lacking from the evidence at Karkemiš, the rich corpus of 10th-9th century BCE texts and images from the site, with support from the 12th-11th century BCE evidence from Malatya, seems to suggest a continuous and perhaps simultaneous development of cultic and political institutions within the region all the way down to the 8th century BCE.

Architectural evidence is less abundant for Malatya, however, the recent excavations at Arslantepe have demonstrated a late 13th century BCE, post-Hittite phase of monumental building activity illustrating an immediate recovery and resilience by the local elites following a destruction process; this rapid recuperation was then followed by a period of decline around the early 12th century BCE (Frangipane *et al.* 2019-2020: 78; Manuelli *et al.* 2021: 888-889; Manuelli 2019: 163; Manuelli 2020: 603). The period from the 12th to 11th century BCE, however, reflects a legacy of Hittite traditions and a close association with Karkemiš and Aleppo through a shared continuity of cultic institutions (Frangipane *et al.* 2019-2020: 80). This is demonstrated in the iconographic repertoire concerning depictions of the Storm-god, in particular, along with a local emphasis on libation scenes characteristic of previous Hittite imagery and cult practices (Manuelli, Mori 2016: 211-213, 226-228). Divine iconography at Malatya demonstrates a persistence of Late Bronze Age Hittite traditions, but is also reflective of regional Early Iron Age trends. The local Storm-gods are still depicted in short belted-tunics, wearing a horned helm, and often standing in a smiting posture, but the style of the horned helms with ridged horns across the lower edge and with an internal element that resembles the Anatolian Hieroglyph SOL is a feature shared only with the Palastinean reliefs at Aleppo, and the Storm-god mounting a bull-drawn eagle-chariot is similarly directly paralleled in a 10th century BCE relief in the temple of the Storm-god at Aleppo (ALEPPO 4; Fig. 5).4 Specifically Karkemišean influence can be seen in the pairing of Kubaba and Karhuha on one relief (MALATYA 13), however, the spelling of the latter god's name, (DEUS.CERVUS2) *kar-hu-ha-sa*, is suggestive of a translation from the Luwian Stag-god, Runtiya (Hawkins 2000: 328-329), and the additional pairing of Hebat and Šarruma on several nearby monuments is indicative of a strong Hurro-Luwian tradition, and reflective of interactions with the regions formerly controlled by Kizzuwadna and Aleppo, that persists well into the Iron Age (DARENDE; GÜRÜN; Hutter 2017: 116; Hutter 2021: 290-299).5

These cultic institutions continue through the 10th century BCE, when monumental building XLVI was constructed with installations suggesting a cultic or ceremonial function and architectural *comparanda* in the southern Levant (Frangipane *et al.* 2019-2020: 81-84), and when the last phase of Early Iron Age sculptural reliefs was carved and set up

<sup>4</sup> Likewise, the image of the Sun-god(dess) on MALATYA 14 is topped by a winged-SOL sign, which may be the same iconograph that rests above the female figure on the Palastinean MEHARDE stele dated *c.* 900 BCE.

<sup>5</sup> This frequent pairing in Malatya may provide yet another link between the cultic institutions of the Upper Euphrates and Aleppo, where Talmi-Šarruma, a Late Bronze Age king and descendant of the Hittite king Suppiluliuma, built a joint temple for Hebat and Šarruma (ALEPPO 1).

Fig. 5 Storm-god of Aleppo and Taita I of Palastina (a: ALEPPO 5+6; Hawkins 2011: 39 fig. 3 [photo by K. Kohlmeyer]), compared with the Storm-god and PUGNUS-mili of Malatya (b: Orthmann 1971: Malatya A/5a; Bilgin); Storm-god mounting bulldrawn eagle chariot at Malatya (c: Orthmann 1971: Malatya A/11; Bilgin), and Aleppo (d: ALEPPO 4; Hawkins 2011: 39 fig. 2 [photo by K. Kohlmeyer]).

before a destruction around the end of the 10th century BCE (Manuelli, Mori 2016: 224- 226). This phase coincides also with an array of silos that suggest a revival of administrative practices characteristic of Hittite imperial traditions, a feature also seen elsewhere in the Syro-Anatolian region at this time (Frangipane *et al.* 2019-2020: 81-84; Castellano 2018; Heffron *et al.* 2017: 134-142; Blaylock 2009: 102). Reuse of Early Iron Age relief orthostats in the newly built Lion's Gate during the subsequent period, along with the construction of the so-called Pillared Hall (Frangipane *et al.* 2019-2020: 87-95), illustrates a resilience of cultic institutions into the 8th century BCE, only ending with Assyrian occupation that included the destruction of the Pillared Hall, the burial of a royal statue in front of the gate, and construction of new, monumental buildings (Frangipane *et al.* 2019-2020: 89-97).

Institutional continuity is likewise evident in the visual and textual assemblage of Karkemiš, but only confidently from the 10th century BCE onward, thus largely coinciding with the Middle Iron Age (Hawkins 2000: 73-223; Marchetti, Peker 2019-2020: 278). The earliest inscribed monuments yet known from post-Hittite Karkemiš are two non-figural stelae dedicated to Ura-Tarhunza, Great King and Hero of Karkemiš, dated to the early 10th century BCE and perhaps recounting events of the 11th century (Dinçol *et al.* 2014: KH.11.O.400; KARKEMIŠ A4b); the titulary mirrors those associated with the early 12th century BCE Kuzi-Teššub, especially in the Malatyan context (GÜRÜN; KÖTÜKALE), suggesting resilience of the political institutions of the kingdom, or at the very least an informed revival (Hawkins 1988). The same inscriptions illustrate the persistence of two major cults – those of the Storm-god, presumably Tarhunza as suggested by the theophoric element in the king's name, and Kubaba, whose priest commissioned the latter of the two monuments and erected it in the temple area – and the emergence and coexistence of a second set of political titles alongside Great King and Hero – those of the Country-Lord and, from the 10th century BCE, the *tarwani* or *tarrawanni*, translated as Ruler or 'the just one', respectively.6 While the latter titles (and associated dynasty) appear to have won out in the succeeding period, a similarly complex political dichotomy and transition can be recognized in the 8th century BCE with the relationship between King Kamani and his vizier Yariri, who acted as regent (*tarwani/ tarrawanni*) during the king's youth (e.g., KARKEMIŠ A6, A7, A15b), and whose descendants appear to have succeeded to the throne, rather than Kamani's own (ADANA 1).

This resilience at Karkemiš is supported by architectural continuity and the embellishment of the monumental space around the lower city Temples A and B, which evidently served as a theatre for public rituals during the 10th and 9th century BCE (Gilibert 2011: 97-131; Gilibert 2014; Marchetti, Peker in Marchetti (ed.) 2019-2020: 278-279). The reliefs from this period include typically Hittite mythological creatures, including bull-men with horned helms, as well as representations of deities that were central to the Late Bronze Age cults of Hittite Karkemiš, such as the Storm-god and Kubaba, and perhaps also Ištar-Šauška and Maliya (Orthmann 1971: 276-277; Hutter 2017: 114-115; Hutter 2021: 300-311). The iconography from these reliefs attests to the persistence of traditional Hittite features, still used, or co-opted, by later Iron Age rulers of Karkemiš as a legitimating force, especially in the connection between the two primary deities and Karkemišean kingship. The worship of the Storm-god of the Vineyard – mostly a micro-regional trend centred around the northeast Mediterranean and often in conjunction with definitions of local kingship – is evident also at Karkemiš,

<sup>6</sup> For a discussion of these titles, see Hawkins 1988; 1995; 2000: 73-79. For *tarwani*, in particular, see Giusfredi 2009. Most recently, Melchert (2019) has offered a convincing alternative, reading the latter term as a substantive adjective, *tarrawanni*. For more on the debate, see also Oreshko 2018: 111-112.

where the deity is represented on a stone stele (ADANA 1) and probably in a bronze figurine (Marchetti 2014: 310-315; KH.11.O.516),7 and may have provided a new means of defining Karkemišean kingship during the 8th century BCE (Lovejoy 2022; Lovejoy forthcoming; Matessi, Lovejoy forthcoming). These local cult structures and sculptures only fell out of use after the conquest of Sargon II in 717 BCE during the period of Assyrian occupation of the site, when both temples were abandoned and replaced, and a new Neo-Assyrian palace was built (Cavriani *et al.* in Marchetti (ed.) 2019-2020: 284). Until that time, it appears that the Upper Euphrates region developed continuously, demonstrating substantial resilience in both cultic and political institutions.

# 2.2. Reorganization in South-Central Anatolia

We have limited information on the LBA-IA transition in south-central Anatolia because we lack archaeological evidence, particularly for the LBA. LBA occupations have been excavated at Ovaören, Niğde-Kınık Höyük, and Porsuk-Zeyve Höyük. Only at Porsuk-Zeyve Höyük has an extensive exposure of LBA levels been produced, but also there it only included the western and eastern slopes of the mound. In none of the three sites have cultic buildings been so far uncovered for the LBA. Moreover, while continuity of occupation throughout the transition seems to characterise the mound of Kınık Höyük, and likely also of Ovaören (Şenyurt, Akçay 2018; d'Alfonso 2020a), a recent reassessment of the dating and stratigraphy of Porsuk-Zeyve Höyük indicates the presence of a significant hiatus in occupation, even though its length still remains unclear (Beyer 2010; 2015). By contrast, many fruitful hints derive from surveys, particularly those covering non-urban sites. Four landscape cultic places have been investigated in detail in relation to the question of the continuity of cults in the LBA-IA transition. These are the Yalburt pond, the Throne peak at Kızıldağ, the Ivriz spring, and the figurative stele with its base at the Tavşantepe pass in the Altunhisar valley.

The Yalburt pond close to the city of Ilgin in the Konya Plain has been at the centre of a micro-regional survey investigating the occupational and environmental change and the archaeological remains from prehistory to the often-problematic embedment in the contemporary landscape. One of the most relevant results is the evidence for continuity of occupation in the area around the pond of Yalburt before and after its monumentalisation. Harmanşah *et al.* have used this evidence in support of the agency of the local community in the definition of the cultic landscape and the stories animating it. They read the monumentalisation of the pond defined by squared and top-moulded limestone blocks bearing the inscriptions of the military deeds of the Hittite Great King Tuthaliya IV as the result of the appropriation of a cultic, collective space of a local, peripheral community by an external political authority (Harmanşah *et al.* 2022, with reference therein).

The case of the Throne and the other cultic installations at Kızıldag has some similarities with the one of Yalburt. In the recent scholarship, a dating of all monumental features at this site to the 8th century has been suggested (Goedegebuure *et al.* 2020; Massa, Osborne 2022, with previous literature). While this reconstruction is possible if one imagines the presence of copies of archaic monuments, we follow here the recon-

<sup>7</sup> While the bronze figurine appears to hold in one hand a bunch of grapes and stalk of grain – a motif especially common in the imagery of Gurgum and Sam'al – Marchetti prefers to interpret the object as a dagger, perhaps referring to grapes in its hilt, due to the absence of this deity elsewhere at Karkemiš (Marchetti 2014: 311, especially n. 7). It should be noted, however, that development of this deity likely involved Karkemiš, where its predecessor – the Storm-god with subordinate Grapeand Grain-gods – is well attested (Lovejoy forthcoming; Matessi, Lovejoy forthcoming).

struction maintaining that there are monumental Anatolian Hieroglyphic inscriptions, and figurative reliefs and altars with offering cups carved in the rock attesting cultic activities at the site at different times from the end of the LBA into the IA (d'Alfonso, Pedrinazzi 2021; Adiego 2021; Hawkins, Weeden 2021).8

It is unclear in both cases whether the cults continued uninterruptedly, or there was some hiatus, if not in the memory of the local populations, in cultic practice at these sacred places. Equally important is to define whether there is continuity in the worshipped gods, and therefore in the institutions they were associated with, or a transformation of the cults while maintaining the sacred places. There are no clear clues to suggest a continuity of cults at Yalburt, or more precisely no interest in claiming any continuity with the imperial patronage by the new authorities in the following periods. In this sense, Yalburt may be defined as a place for the reorganisation of the cults.

By contrast, following the traditional dating of the monuments, the production of new IA inscriptions and the engraved relief of the ruler on the throne at Kızıldağ suggest a desire of reinforcing the previous model of patronage of the LBA in the new period. The selection of archaic models of inscriptions otherwise not attested in the IA are indicative of a desire to enhance the legacy of a former imperial authority, and to use this legacy in support of the legitimation of the later one. Similarly, the reference to traditionally LBA deities with the prominence given to the Storm-god and the Storm-god of Heaven is indicative of the preservation of the hierarchy in the pantheon as known in Hittite Anatolia. Lastly, the model of an inscription topped by the winged sun-disc without association with other figurative reliefs representing a deity is to my knowledge only produced by Great King Tuthaliya IV in the late 13th century (Aro 2022: 568-570), and afterward by Great King Hartapus at Kızıldağ and Great King Ura-Tarhunzas at Karkemiš (Dınçol *et al.* 2014, with reference to the previously found KARKEMIS 4b inscription) (Fig. 6). In both latter cases, the use of the winged sun-disc specifically refers to the sacred dimension of Great Kingship as defined in the Hittite empire. As in Yalburt, we suggest that the sacred place was active during the LBA-IA transition, but in this case, there are strong clues for a continuity in the association between the sacred place, cultic institutions, and the legitimation of political authority.

The situation at the Ivriz spring and at the pass of Tavşantepe is still a different one. No LBA investment in the monumentalisation of these landscape cultic places has been found, even though there is evidence for frequentation and cultic activity for this period (Maner 2017). During the IA two distinct rock reliefs have been carved. They represent different cultic scenes, in different styles and likely dating to different periods within the IA. The well-known relief of king Warpalawa worshipping the Storm-god has been for a long time interpreted as a sign of continuity with the cult of the Storm-god in south Cappadocia attested during the Hittite empire. In particular, the cuneiform sources give due importance to the Storm-god of Tuwanuwa, and the Storm-god represented at the Ivriz spring is indeed associated with Warpalawa, the king of Tuwana, likely the first millennium form of the name Tuwanuwa (Mora 2010; Mora, Balatti 2012). In fact, the Storm-god of Ivriz is a Storm-god of the Vineyard, as evident in the association with the grapevine in iconography and the way he is explicitly named in the BOR 1 inscription (§4, (VITIS)*tuwarsasin* (DEUS)*Tarhunzan*, acc. s.). While the Ivriz 1 relief is only one of several reliefs of the Storm-god uncovered all around south Cappadocia, thus offering a glimpse of the spread of this cult in the region, it is worth noting that they are all

<sup>8</sup> The relevant presence of cultic activities also in the Hellenistic and Byzantine periods is not discussed, as it is not relevant to this paper.

a

Fig. 6 The KIZILDAĞ 3 rock-engraved inscription (a: Hawkins 2000: Pl. 237), compared with the stele of Tuthaliya IV from Boğazköy (b: BOĞAZKÖY 18; Bilgin).

contemporary, dating to the end of the 8th century BCE, all exclusively representing a Storm-god of the Vineyard, and directly associated with the two kings of Tuwana so far attested, namely Warpalawas and Muhawaranis. N. Lovejoy has recently suggested that the origin of this new aspect of the cult may be exogenous and may have been introduced in south Cappadocia from the northern Levant where it developed out of a tradition attested first at Ugarit and then in the kingdom of Palastina (Lovejoy forthcoming). So, in a way, the existence of more temple institutions of the Storm-god in the region during the Hittite period would make the new cult particularly easy to accept by local communities. At the same time, the cult has clear aspects of innovation for this region, possibly also associated with the transformation of the landscape, characterised by a more intensive production of grapes and bread wheat (Castellano 2021).

The stele of Tavşantepe shows the image of a female goddess seated on a throne over the back of a crouched lion. While this has often been interpreted as a representation of Kubaba, whose cult in the region is known thanks to the inscription of Bulgarmaden, A. Lanaro has suggested that she could in fact represent one of the southern Cappadocian goddesses associated with the agricultural productivity of the land well known from the cuneiform religious texts from Hattusa dealing with cultic activities and institutions of southern Cappadocia. In opposition to the many standardised representations of the Storm-god of the Vineyard, the Tavşantepe stele shows no direct or mediated Assyrian influence, and Lanaro suggested for it a dating in the beginning of the 1st millennium BCE (Lanaro 2015). In its crude style, the Tavşantepe stele is similar to the Ivriz 2 relief, representing an offering scene which has the best parallels in the Hittite and immediately post-Hittite period. D'Alfonso has recently suggested that Tavşantepe and Ivriz 2 represent a first reorganization of local cults in south Cappadocia after the fall of the Hittite empire, whereas the cult of the Storm-god of the Vineyard represents an innovation introduced from the northern Levant through Cilicia distinctly associated with the last two kings of Tuwana of the late 8th century (d'Alfonso 2020b).

# 2.3. Transformation in North-Central Anatolia

North-Central Anatolia is the region in which the transition from the LBA to the IA took the most evident and profound transformation in political organisation and social complexity. While the EIA and MIA occupation has traditionally received less attention than the earlier Hittite period, today excavations at a number of sites allow for a regional overview of the transition. The whole reconsideration of the transition in this period is very much indebted to the excavation led by Dr. Seeher at Büyükkaya-Boğazköy. The study of architecture and material culture, in particular ceramics, could identify a profound *caesura* between the late imperial occupation and the following EIA phases, characterised by an immediate change in architectural technique and space organisation at the site, as well as a progressive abandonment of the Hittite ceramic shapes and production technology, moving to new, simplified shapes, handmade production, and adoption of a few, well identifiable, red-painted decorative motifs under the rims of selected closed and open vessels (Genz 2004; Seeher 2018). None of these EIA architectural features (mostly one-room, squared pit-houses, as well as middle-sized pits) could be associated with cultic activities. Recent studies of the metallurgical and agropastoral activity at the site, and at the neighbouring site of Çadır Höyük, seem to imply that the transformation should not be understood as a complete abandonment or reduction of scale of the activities taking place in the region, or a complete isolation from former and new long-distance interaction. They rather show the reorganisation of economic activities within a profoundly transformed political and social scenario (Lehner 2017; Ross *et al.* 2019).

Fig. 7 The statue of 'Matar' from the main room in the Büyükkale I fortification gate at Boğazköy (Wikimedia Commons, https://commons.wikimedia.org/wiki/File:Limestone\_statue\_of\_ Cybele,\_the\_Phrygian\_Mother\_Earth\_goddess,\_flanked\_by\_two\_male\_figures,\_discovered\_in\_Bo%C4%9Fazk%C3%B6y,\_6th\_century\_BC,\_Museum\_of\_Anatolian\_Civilizations,\_Ankara\_(29652460984).jpg).

The early first-millennium occupation at Büyükkaya equally did not entail any public building to be associated with cultic activity of the local community. This Büyükkaya occupation phase has been recognised as either slightly earlier, or even contemporary with the reoccupation of the citadel of the site of Boğazköy, namely the Büyükkale II occupation phase. Excavated at a very early stage, the stratigraphy of the IA occupation of Büyükkale is known to a much lower degree of depth than the one of the Hittite and pre-Hittite periods. This notwithstanding, after a hiatus of some 300 years, the citadel is characterised by single room, squared buildings of roughly the same dimensions, and the adoption of Dark Monochrome Geometric Painted (DMGP) ware, for the most part wheel-made. The adoption of the DMGP ware, and in particular of its Alişar-IV subset, for wine drinking is indicative of the participation in the dynamics of multi-centred, interrelated canton states whose earlier examples are emerging in south-central Anatolia during the 10th century BCE. Within this occupation phase, again, no evidence of cultic space or cultic activity could be singled out. The earliest evidence of non-private cult comes in fact from the Büyükkale I occupation phase, dated to the 7th century BCE. This consists of the niche with the cultic statue of Matar/ Kybele occupying a corner of a room of the gate providing access to the citadel (Neve 1982: 152-154; Summers 2021: 137-138), as well as a cultic deposit found in the lower town, out of the citadel (Bittel 1970: 139-141). Also in this case, materials are indicative of cultic activities related to the cult of Matar/Kybele (Fig. 7). It is worth underscoring that there is no trace indicative of the existence of the worshipping of the Storm-god 194 Lorenzo d'Alfonso, Nathan Lovejoy

Map 1. Sites with non-private cultic space discussed in the paper, divided between LBA, and IA sites: 1. Nerikka - Oymaağaç Höyük; 2. Yazılıkaya; 3. Hattusa - Boğazköy; 4. Zippalanda - Uşaklı Höyük; 5. Sarissa - Kuşaklı Höyük; 6. *huwasi* near Kuşaklı; 7. Altınyayla; 8. Yalburt; 9. Malatya - Arslantepe; 10. Fıraktın; 11. Imamkulu; 12. Tavşantepe; 13. Kınık Höyük; 14. Emirğazi; 15. Eflatun Pınar; 16. Kızıldağ; 17. Ivriz; 18. Domuztepe; 19. Azatiwataya - Karatepe; 20. Tatarlı Höyük; 21. Zincirli Höyük; 22. Sirkeli Höyük; 23. Adana; 24. Tarsa - Gözlükale; 25. Karkamiš; 26. Cağdın; 27. Kilise Tepe; 28. Tell 'ain Dara; 29. Arsuz; 30. Kinalua - Tell Tayınat; 31. Alalaḫ - Tell Atçana; 32. Halpa - Aleppo; 33. Karaburun/Hatay; 34. Emar - Tell Meskene; 35. Tell Afis; 36. Ugarit - Ras Shamra; 37. Tell Tweini; 38. Hama; (Prepared by L. d'Alfonso and L. Castellano with software ArcGIS).

or the Sun deities that had a prominent position in religious, institutional, economic, and political functioning of the Hittite empire. The evidence from the site hints at the complete lack of non-private cults from the early 12th century down to the 7th century BCE. Together with the lack of contexts associated with political authority, the lack of evidence for cultic institutions during the EIA and the MIA, and the emergence of cults in town showing no legacy of the former LBA religious institutions are the strongest evidence of the profound transformation of social and political complexity at the site of the former capital of the empire.

A brief overview of other sites excavated in the region provides clues for extending this reconstruction to the whole region of NCA. Besides Boğazköy, excavated sites with well-documented evidence of the LBA-IA transition are Çadır Höyük, Uşaklı Höyük, Maşat Höyük, and towards the Pontic region Akalan, İkiztepe, Oluz Höyük and Oymaağaç Höyük (see Yılmaz contribution to Ökse, Czichon, Yılmaz 2021). The monumental temple buildings at both Oymaağaç Höyük and Uşaklı Höyük were not reconstructed or reused during the Early or Middle Iron Ages. No evidence of non-private cultic space emerged from the investigations at Maşat, either (Özgüç 1978: 91- 94). In the whole region, there seems to be a profound shift in religion, with little to no evidence of continuity of cultic institutions and cults of the previous Hittite period.

#### 2.4. Reorganization and Transformation in Cilicia

With the end of the Late Bronze Age, the Cilician Plain suffered a widespread disruption to both cultic and political institutions. While the previous period featured several structures representative of Hittite cultic institutions – namely, the aforementioned shrine or open-air sanctuary at Sirkeli Höyük (van den Hout 2002: 89-91; Kozal, Novák 2013: 233), a monumental structure at Tarsus interpreted as a Hittite temple (Goldman 1956: 49-50; Trameri 2020: 440-441), the North-West Building/Stele Building sequence at Kilise Tepe (Bouthillier *et al.* 2014; Blakeney 2017: 46-47), and a possible Hittite temple atop the citadel mound of Tatarlı Höyük (Girginer 2011: 133-135; Girginer, Oyman-Girginer, Akıl 2012: 110-112) – the end of the second and early first millennium is devoid of any certain evidence of temples, sanctuaries, or shrines that might reflect the later cultic institutions of the region. They are clearly evident in the sculptural and inscriptional sources, albeit only from the Middle Iron Age. A number of monumental structures produced throughout the period may have served some cultic function, but they have not produced material or textual evidence to assign them a descriptor any more specific than 'special.'9 While the monumental building projects and extensive urban expansion of Sirkeli Höyük, which occurred without any substantial hiatus of occupation throughout the Iron Age, suggest a rapid re-emergence or continuity of some central authority at the site following the loss of Hittite control, any political institution at the site appears to be wholly local. This appears in stark contrast to the situation in the previous period, where the regional authority of the site was centred around its sacred relevance, especially to the Hittite empire, and a ruler with

<sup>9</sup> Monumental structures on the mound of Sirkeli Höyük dated as early as the mid-late 12th century BCE (Novák 2020: 215-220; Sollee *et al.* 2020: 221-224), the apsidal structure at Tarsus dated between the mid-12th and mid-10th century BCE (Goldman 1963: 3-6; Yalcın 2013: 200; Ünlü 2015: 519-520), the posthole ring or apse at Kilise Tepe dated to the beginning of the 12th century BCE (Postgate, Thomas 2007: 121; Blakeney 2017: 46-47; Heffron *et al.* 2017: 118-120), a 9th century BCE renovation of the Hittite temple at Tatarlı Höyük (Ünal, Girginer 2010: 275; Girginer, Oyman-Girginer 2020).

Fig. 8 Stele of the Storm-god from Domuztepe (Photo by N. Lovejoy).

the title of 'Priest.' This lack of institutional continuity is immediately suggestive of an intensive reorganisation at the Sirkeli Höyük and likely throughout Cilicia as a whole.

Sculptural and textual evidence provides more information pertaining to the cultic institutions of Cilicia, but only for the 9th to perhaps the early 7th century BCE. The earliest evidence, coming from the 9th-early 8th century BCE, includes the fragmentary sculptures of Domuztepe and a stele depicting the Storm-god of the Vineyard and commissioned by a Karkemišean official found near Adana. At Domuztepe, three broken stelae depicting gods, two identifiable as the Storm-god and Kubaba, while the third stands upon a sphinx with its identity unknown, and a stele illustrating a 'tree-of-life' motif beneath a winged sun disc provide the earliest indication for the continuation or resumption of local cultic institutions; from the imagery of Domuztepe alone, the cults could have recovered or evolved from the Aleppine/Kizzuwadnean traditions with little to no outside influence (Fig. 8). However, the Karkemišean stele provides evidence for the introduction of a new hypostasis of the Storm-god in the region. While the citadel of Domuztepe is likely indicative of a new political institution in Cilicia (with a Hurro-Anatolian background), abrupt institutional change is evident in the following century with the Luwo-Phoenician monuments of the later 8th century BCE rulers of Hiyawa (Bossert *et al*. 1950; Alkım 1952: 240-249; Ussishkin 1969; Winter 1979; Çambel, Özyar 2003: 153-156; Sicker-Akman *et al.* 2014).

Whether through some form of migration or persistent interaction, 8th century BCE Cilicia came to be characterised by the use of the Phoenician language and script and its impact in the previously Hurro-Luwian setting. The rest of the monuments from the region date to the second half of the century, or shortly thereafter, and are all products of two rulers, Awarika and Azatiwada. The inscriptions of these rulers all include Phoenician texts, and each ruler also produced a Luwian-Phoenician bilingual.10 In both instances, these texts are found on statues representing a local hypostasis of the Storm-god of the Vineyard, possibly inspired by the intrusive stele from Karkemiš, and the figures are identified as both the Luwian Tarhunza and the Phoenician Ba'al (Lovejoy 2022; Matessi, Lovejoy forthcoming). In the KARATEPE inscription, we also find the Luwian Stag-god Runza translated into the Phoenician Resheph of the Goats, and Ea translated into El. The imagery from Karatepe is also striking in comparison to the more traditional Neo-Hittite character seen at nearby, and only slightly earlier, Domuztepe. While the Levantine or eastern Mediterranean character of several elements from the relief orthostats of Karatepe has long been noted,11 the presence of the Egyptian god Bes apparently acting in his characteristic apotropaic role on two reliefs of the North Gate bears emphasis, as it is additional and explicit evidence of (perhaps Cypro-) Phoenician cultic influence at the site (Orthmann 1971: 267-271).12 It should also be noted that there seems to be no attempt by the stoneworkers of Karatepe to revive the traditional Luwian cultic imagery from Domuztepe, except perhaps in the double bull statue base of the Storm-god.

The Storm-god is no longer represented in the guise of a warrior, but stands placidly in a long robe. This is a major change in the representation of the Storm-god, definitively distant from the Hittite, second millennium figurative and symbolic representation. Similarly, the cult of Kubaba disappears entirely, as she is not mentioned in any of the Hiyawan inscriptions and none of the myriad images from Karatepe could be interpreted as hers. In sum, while the wholesale societal disruption that took place during the Early Iron Age caused the dramatic silence in archaeological, art historical, and epigraphic sources pertaining to cults and cultic institutions in the region, the reprise in the 9th century BCE and the introduction of new cults by foreign elites from the regions east of the Amanus and from the Levantine coast led to a complete institutional reform. This reform resulted in a new cosmopolitan conceptualisation of the divine in which the gods most closely tied to Hiyawan kingship bore multiple identities. And while each ruler sought to express their own political identity in unique ways that reflect the changing socio-political landscape of the later 8th century BCE (Lovejoy 2022), the institution of Hiyawan kingship was transformed already through its association with new sources of legitimation. While the Storm-god remained the foundation of sacral kingship, in Hiyawa, it was no longer the warrior god wielding axe and lightning, but a provider god bearing grape and grain.

<sup>10</sup> While the İNCİRLİ inscription is usually called a trilingual (Kaufman 2007), this classification is debatable. While the Phoenician text appears to be primary, the undecipherable Akkadian may be a secondary inscription, and the so-called Luwian inscription might simply be the EGO sign initiating 1st person speech.

<sup>11</sup> Egyptianising sphynxes, a breastfeeding woman, Phoenician ship, etc. (Orthmann 1986; Özyar 1998; Çambel, Özyar 2003: 131-140; Özyar 2003: 110-112; Özyar 2013: 128-131). For an argument towards Neo-Assyrian influence at Karatepe, see Aro 2014.

<sup>12</sup> For iconography associated with Bes in the eastern Mediterranean, see Culican 1968: 92-98; for a brief description of Bes' primary functions, see Hart 2005: 49-50.

# 2.5. Resilience, Reorganisation, and Transformation in the Northern Levant

The Early Iron Age cultic institutions of the northern Levant demonstrate substantial continuity, unseen elsewhere in the Syro-Anatolian region outside of the Upper Euphrates. The temples of Aleppo and 'Ain Dara survived the Late Bronze Age fragmentation and subsequent vacuum of political powers in the region and even demonstrated the earliest signs of a return to prosperity and stability. The temple of the Storm-god at Aleppo was renovated by a king Taita I of Palastina during the 11th century BCE. These renovations included new relief orthostats, new portal figures, a relief carving of the king facing that of the Storm-god, and several inscriptions. The excavators have also interpreted a reorientation of the space, returning to a direct-axis approach to a new primary divine figure, however, the elaboration of the space around the Storm-god relief with the royal figure and monumental inscription suggests, instead, a continuity of the bent-axis plan (*contra* Kohlmeyer 2009: 197-200; Aro 2010: 5; Hawkins 2011; Kohlmeyer 2012: 64-65, 68-69). Likewise, the similarities in style between the figures of the king and deity suggest an attempt at imitation, albeit with certain choices reflecting new cultural norms, e.g., the lack of curled toes on the ruler. Note the v-neck tunic and pointed kilt, only simplified without decoration for the king; see also the pointed cap of Taita, lacking the horns and the internal SOL sign that mark the divine nature of the Storm-god, besides the short epigraph defining him as the Halabean hypostasis of the deity (ALEPPO 5).13 These differences suggest an awareness of the significance of certain iconography (i.e., the divine helm) and an explicit effort to distinguish the new ruler in a position lower than the god, despite the peculiar proportional relationship between the two figures. Additionally important, the king's inscription (ALEPPO 6) describes a set of ritual sacrifices that should be made to the Halabean Storm-god by men of various standing, from king to commoner, whenever they should come to the temple. This proclamation directly ties Taita's kingship to the cultic institution, providing a foundation for his power at a time when we know of no Palastinean palace or other governmental structure. Taita I's other inscription from the temple (ALEPPO 7) provides another important connection. While the text is incredibly fragmentary, a possible reference to Kubaba, seemingly in the context of an interaction with Karkemiš, indicates a continued or renewed relationship between northern Syria and the Upper Euphrates, providing a clear venue for cultural exchange in the cultic context (Singer 2012; Weeden 2013; 2015; Younger 2016: 123-134; Simon 2019: 136-138, esp. n. 55).

The temple at 'Ain Dara, dedicated either to the Storm-god or Ištar-Šauška, was similarly renovated with artistic comparison suggesting the work was also commissioned by Taita I (Abu Assaf 1990; Novák 2012: 48-49). At Tell Afis, a new, probably *in antis*, temple to the Storm-god was built during the 11th century BCE and renovated during the early 10th century BCE. Finds within the space include a cylinder seal depicting the Storm-god and several vessels and other objects associated with ritual behavior (Soldi 2009: 106-108; Mazzoni 2012: 24-26; Mazzoni 2014: 47-51; Mazzoni 2019: 311-312, 318-319; Fig. 9). In contrast to Cilicia, this Early Iron Age new construction and the renovations at each excavated temple of the region indicate that local cultic institutions not only survived the collapse of the Late Bronze Age political systems, but returned to prominence soon after – long before political institutions or infrastructures re-emerged in the region.

<sup>13</sup> See also Section 2.2 for arguments based on similarities between the Aleppo Storm-god relief and the Storm-god reliefs at Malatya.

RULERSHIP AND THE GODS 199

In the subsequent Middle Iron Age, the cultic landscape of the northern Levant changed considerably, no longer characterized by a small number of intra-regional cult centres, but rather reorganised around local temples within major urban spaces. The temples of the previous period each experienced phases of renovation and transformation before most were eventually destroyed, abandoned, or replaced. These processes are similarly paralleled in the political institutions of the region, with the apparent transplantation of the Palastinean monumental centre to Tell Tayinat and the subsequent or simultaneous rise of surrounding polities with new socio-political structures, such as those of Zincirli, Bit-Agusi, and Hamath and Luash.

The temple at Aleppo was renovated for a final time around 900 BCE with the addition of several new relief orthostats before its subsequent conflagration and abandonment. The temple at 'Ain Dara also appears to have been renovated in the 10th century

Fig. 9 Tell Afis temple of the Storm-god and Storm-god seal (Mazzoni 2014: 45 fig. 3, 46 fig. 4, 51 fig. 16, courtesy of the Tell Afis Archaeological Project).

and somewhere in the 9th-mid-8th century BCE, based on stylistic analysis of the relief orthostats. The temple of the Storm-god at Tell Afis likely remained in use until around the late 8th-early 7th century BCE; at this time, a series of two new temples were built over it, the last of which bears substantial Neo-Assyrian influences, and may be attributed to Assyrian occupation or administration of the site (Mazzoni 2012: 25-35; Mazzoni 2014: 45-47; Cecchini 2014: 58-61). There is also possibly a mid-10th to mid-9th century BCE temple located within the domestic space of Tell Afis, suggested by what Mazzoni has interpreted as an altar, however, this interpretation is doubted by Venturi on the basis of the poor preservation of the structure's plan and finds (Mazzoni 1998: 165-166; Venturi 2007: 187; Mazzoni 2012: 27-29; Venturi 2020: 37-40). A new temple district was also constructed at Tell Tweini between the 10th and mid-9th century BCE, including what is interpreted as an *in antis* temple structure along with a walled, cobblestone-paved courtyard and plastered ashlar platform with a carved hole suggestive of drainage for libations or sacrifices. The excavators interpret the sacred space as Phoenician based on architectural and material *comparanda* at Sarepta and Enkomi, though perhaps these connections would more precisely imply a Cypro-Phoenician cultural context; in any case, it is unclear to which deity or cult the space may have been dedicated (Bretschneider *et al.* 2000: 87-96; al-Maqdissi *et al.* 2007: 62-63; Bretschneider *et al.* 2008). During the 9th-8th century BCE, an Assyrian open-air sanctuary was created at Karabur, located about 25 km south of Antakya, thus not far from the Amuq Plain. This sanctuary comprises conical granite outcroppings spread over an area of about 100 m, including four such outcroppings with reliefs depicting divine figures and symbols, and at least one worshiper (Taşyürek 1975: 172-180). The eroded state of the reliefs and lack of inscriptions do not permit a more precise dating of the sacred space, and the variety of symbols and iconography suggest that it may have served a multitude of deities, perhaps including both those of Syro-Levantine and Assyrian pantheons.

Dating from the 10th to early 9th century BCE are also a set of important sculptural monuments from the northern Levant, all attributed to the kingdom of Palastina.14 It is also important to consider that this period represents the first monumentalisation and urbanisation of the Palastinean capital, Kunulua, at Tell Tayinat in the Amuq Plain. Two funerary stelae invoking a goddess known as the 'Divine Queen of the Land,' one apparently depicting the goddess standing on a lion, wearing a 'Hathor-headdress,' and surmounted by a winged sun seemingly represented by the stylised Anatolian Hieroglyph SOL in much the same way as on MALATYA 14, were erected by Taita (probably II) and his wife(?) Kupapiya; beside the deity, standing upon the lion's head, is a smaller figure appearing to represent the king, quite similar to the way the Ugaritic king is represented on the Late Bronze Age Ba'al with Thunderbolt stele (SHEIZAR; MEHARDE). Two stelae of Suppliliuma I of Palastina depict the Storm-god guiding a smaller royal figure by the hand and holding a lightning trident before him. The king in the relief carries a bunch of grapes and stalk of grain in his hands, seeming to represent the Grain- and Wine-gods invoked in the accompanying inscription (ARSUZ 1 and 2). This pair of deities appears to be an innovation of the period, appearing also at Karkemiš,

<sup>14</sup> While there are disagreements between the specific chronology of several of these monuments (Giusfredi 2018, with references therein), they will be irrelevant for the current argument. The monuments of SHEIZAR, MEHARDE, ARSUZ 1 and 2, and ALEPPO 4 with associated reliefs will here be considered as a group without an attempt at ordering them. Likewise, the disputed association of SHEIZAR and MEHARDE with the region of Hama will not be considered. Together, these monuments are examined as products of the Palastinean 'body politic' within the 10th to early 9th centuries BCE northern Levant.

and seemingly derived from the Ugaritic mythological context where the messengers of the Storm-god Ba'al are known as Gapn and Ugar, or Vine and Field (Smith, Pitard 2009: 222-223, especially n. 39; Lovejoy forthcoming). The last cultic monuments dated to this period come once again from the temple of the Storm-god at Aleppo, where the central chamber was renovated with several new orthostats depicting mythological and divine figures. Most prominent among these is a single long orthostat at the centre of the pedestal wall bearing the image of the Anatolian Stag-god Runtiya following a god driving a bull-drawn eagle-chariot and identified as the 'Divine Mace,' perhaps a hypostasis of the Storm-god (ALEPPO 4). Interestingly, these deities resemble the 11th century BCE image of the ruler Taita I more than the 13th century BCE image of the Storm-god from the same room; their toes are not curled, their beard is short and hair a single mass, their tunic and skirt are plain, and especially noteworthy is their pointed cap with ribbed external surface: though the caps bear a single set of small horns, they lack the internal SOL of the Storm-god. While this could be an intentional difference aimed at separating the deities in hierarchical position, it may also reflect a loss of understanding of the significant iconograph that represents the solarisation of the Storm-god.

At Tell Tayinat, two temples *in antis* were constructed between the mid-9th and mid-8th century BCE, possibly serving as the divine residences of the Storm-god and his consort, as suggested already by Harrison (2012: 19). Both of the Palastinean temples at Tell Tayinat were renovated during the period of Assyrian occupation in the late 8th to early 7th century BCE, and a large platform was built nearby, expanding what Harrison describes as a 'sacred precinct,' reflecting Assyrian practices (Harrison 2009: 184-186; Harrison 2011b: 35-36; Harrison 2012; Harrison, Osborne 2012; Fig. 10). These renovations came with a reorientation to Mesopotamian deities, at least for the smaller of the two temples, which became a temple of Nabu; the tablet collection kept in the *adyton* of the temple finds *comparanda* in the temple of Nabu at Nimrud, where eight copies of Esarhaddon's loyalty oath were found (Harrison 2012: 16). While new occupants have also been proposed for the larger temple, its greater size would have likely resulted in it remaining the primary cultic structure in the area, probably still housing the Storm-god as the head of the local pantheon (Petrovich 2016: 110-141). The intentional destruction and deposition of earlier monumental statuary, perhaps the focus of dynastic or ancestor cults, and the prominence of the Assyrian loyalty oath tablet within the new temple supports a dramatic reimagining of cult practice at Tell Tayinat (Harrison 2011a; Harrison 2011b: 34; Lauinger 2011; Harrison 2012: 16; Harrison *et al.* 2018; Denel, Harrison 2018: 369-370).15

Around Aleppo, while the major temple was no longer in use, the mid-9th to 8th century BCE was characterised by the emergence of Bit-Agusi, perhaps enabled by the fall of the supra-regional cultic institution. During this period, new cultic institutions are suggested by several stelae bearing Aramaic inscriptions: one depicts the striding figure of Melqart wielding an axe or cudgel, commissioned by Bir-Hadad, king of Aram (KAI

<sup>15</sup> Inscribed monuments from the Amuq Plain are limited during this period and mostly fragmentary. From Tell Tayinat itself is a fragmentary statue base whose inscription refers to several deities including Ea the King, the Grain- and Wine-gods, the Sun-god, Tarhunza, Runtiya, and Tamukina, perhaps the local form of the Mesopotamian goddess Damqina, the wife of Ea (TELL TAYINAT 2). A broken statue base from Jisr el Hadid describes offerings for Tarhunza (JISR EL HADID 4), and a building block from Tuleil refers to Kubaba and the Harranean Moon-god (TULEIL 2). The lower half of a small statue of a robed figure discovered near the village of Hatay describes offerings made to the Divine Queen of the Land, presumably the figure represented by the statue itself (KIRÇOĞLU).

201; *CoS* 2.33), while a set of three or four stelae recording a treaty between local kings includes a long list of divine witnesses, beginning with Mesopotamian deities followed by local Syro-Levantine deities and forces of nature (KAI 222-224; *CoS* 2.82).16 Similarly, the region around Tell Afis appears to have fallen under the control of Zakkur, king of Hamath and Luash, around the early 8th century BCE. An Aramaic inscribed stele discovered at the site is dedicated to the god El-Wer and credits Ba'al Shamem as the legitimating force behind Zakkur's rule of the city of Hazrach, likely Tell Afis (KAI 202; *CoS* 2.35).

It is clear that if the Early Iron Age was a period of continuity and resilience of cultic institutions from the Late Bronze Age, the Middle Iron Age was rather a period of reorganisation followed by substantial transformation and diversification of the cultic landscape in the northern Levant. At this later stage, the central position of the Stormgod in the cultic institutions of the region appears less clear from the evidence of sacred constructed space, with greater plurality of temples across the region. This is not evident in the textual sources, however, as the Storm-god persists as the head of most panthea; nevertheless, with the intrusion of several foreign influences during the 9th to 7th century BCE, an increased diversity of cultic institutions is reflected in the evidence taken together. The intensifying Assyrianization towards the end of the period likewise appears to have overshadowed local cultic institutions or at least reframed them within the Assyrian worldview.

In summary, for the northern Levant, the persistence in regional temples with their associated cults provided an anchor on which a new political structure could ground itself, thus perpetuating previous traditions associated with kingship and its connection to the divine. Palastina's initial political stability during the first two centuries of the first millennium is likely the result of this earlier cultic continuity, perhaps demonstrating a long process of political revival grounded upon the resilience of the major temples of the region. However, what is missed in this assessment is the absence of local institutions associated with the NW Semitic-speaking populations following the fall of Ugarit, immediately to the south, where several important temples were never rebuilt. This loss of built sacred spaces might have left communities needing a way to materialize their beliefs or participate in the now more ephemeral cults of the region, perhaps providing an impetus for cultic innovation, e.g., the proto-Storm-god of the Vineyard on the ARSUZ stelae (Lovejoy forthcoming) or the Divine Queen of the Land (Lovejoy, Matessi forthcoming), as institutions merged and produced new outcomes. So, while Palastina begins as a prime example of cultic resilience resulting in political revival and conservatism, the lack of regeneration of NW Semitic cults or the absence of an outlet for religious beliefs of NW Semitic populations may have applied pressure to the new Palastinean power, which had exclusively sponsored and based its legitimacy on cultic institutions previously patronized by the Hittite empire, eventually leading to transformation.

#### 3. Conclusions

This paper has aimed to demonstrate that cults and cultic institutions are a crucial element for the understanding of the processes producing different regional outcomes after the fall of the Hittite empire. In a structuralist or functionalistic perspective,

<sup>16</sup> '[Assur] and Mullesh…Marduk and Zarpani…Nabu and T[ashmet…Ir and Nus]k…Nergal and Las…Samas and Nur…S[in and Nikkal]…Nikkar and Kadi'ah…Rahbah and Adam(?)…[Hadad of A]leppo…Sibitti…El and 'Elyan…Hea[van and Earth…(the) A]byss and (the) Springs…Day and Night – all the god[s of KTK and the gods of Ar]pad (are) witnesses….'

cultic institutions can be seen as a result of specialisation and political complexity. In our emic, contextual, diachronic, and dynamic model, cults are normative cosmic forces defining tempo and worldview of ancient societies, and cultic institutions are a well-identified physical space defined by purity, charged with real and symbolic value, led by specialists whose competence is recognised by the community. Instead of being a by-product of political complexity, they are a driving force behind the power dynamics because they are perceived as such in a bottom-up perspective, but also often by main political actors in search of legitimation of their power.

The land of Hatti was defined by its temple institutions in urban contexts and sanctuaries in the landscape. Temple space and the organisation of cultic ceremonies within the temple and in urban and extra-urban processions were defined by precise norms preserved in the archives of the Hittite capital. When northern Syria became part of the Hittite empire, local cultic space, organisation, and ritual ceremonies – all expressions of local communities – persisted in spite of the impact of the new hegemonic power. In fact, there is growing evidence for the deliberate choice by the Hittites of maintaining the local cults and, in fact, patronising them as a strategy to gain the support of the local population. As for the built environment of the temples, this caused the continuity in the local organisation of architectural space, however, with the addition of Hittite monumental figurative or graphic art, often directly hinting at the Hittite personality acting as new sponsor of the institution.

In the model mentioned above, a resilient attitude of a large community in times of distress results in a form of continuity of the cults representing the normative, cosmic forces defining its group identity. In the Upper Euphrates, there appears to be a simultaneous resilience of cultic and political institutions, which resulted in a continuous process of development within the region. The two major centres seem to have facilitated this continuity in each other, both demonstrating resilience even after periods of decline, destruction, or abandonment.

While processes of resilience, reorganisation, or transformation do not often appear to occur simultaneously, the situation in the northern Levant suggests that new political institutions were, in fact, reliant upon traditional, local cultic institutions. Even when newcomers were able to come to power and brought in new rituals, for example in commensality (Pucci 2019), the legitimacy of their power was dependent on the legitimation provided by the persistence of local cultic institutions and the deliberate decision by this new elite to patronise them, and through them to enter in direct contact with the resilient polities in the Upper Euphrates. We suggest that in this way what James Osborne has defined the 'Syro-Anatolian Culture Complex' was obtained (Osborne 2021).

This was followed by a reorganisation of the cultic landscape and several notable transformations within local cults that coincided with a fragmentation and diversification of both cultic and political institutions with novel, local outcomes. While Palastina itself was reorganized around the Amuq valley, previously important to the Hittite empire, and where even Middle Iron Age kings bore dynastic names referring back to real or imagined ancestors from the great Anatolian power, its peripheral territories were transformed into new local kingdoms with distinct political institutions lacking, for the most part, any major connection with the Late Bronze Age past.

South-central Anatolia, on the other hand, was characterised by processes of reorganisation of both political and cultic institutions. The interruption of evidence for the cult of the Storm-god, so heavily present in the Hittite cuneiform sources describing this region, and instead the relevance of a cult of a female goddess may only be the result of limited investigations, and this may also be the case for the lack of evidence of temples in urban context. Otherwise, they might hint at a meaningful change in social organisation towards the cults and the related group identities in the region. By contrast, the relationship between institutions during the final late 8th century BCE phase of transformation, which resulted in new regional cults and local definitions of sacral kingship, has to be understood as a profound regional innovation based, however, on the revival of the legacy of local cults.

In north-central Anatolia, the core of the previous Hittite empire, an abrupt disappearance of both political and cultic institutions and a period of reduced socio-political complexity is followed only in the 7th century BCE by the emergence of new cults, apparently coinciding with the formations of new polities defined by elements of Phrygian identity – and possibly by direct migration.

Cilicia is a similarly challenging case study, as after the super-regional key role played in the Late Bronze Age, specific evidence of cultic institutions is missing until the 9th century BCE. However, this first evidence appears indicative of a reorganisation of the cultic landscape with some implied resilience in specific cults. This was followed by abrupt changes in the major cultic institutions and deities worshiped in the region, which coincided with striking transformations in political institutions, especially in the definition of Hiyawan kingship.

The unique and varied processes and outcomes of the summarised case studies demonstrate both the importance of such a micro-regionally defined study, as well as the shared coincidence of cultic and political institutional change. It seems evident that cultic continuity coincided with the resilience of political institutions, and changes in the cultic landscape corresponded to political reorganisations or transformations in post-Hittite Anatolia and north Syria.

# Abbreviations


## References


*Mill. BCE)*, Abhandlungen des Deutschen Palästina-Vereins 41, Wiesbaden, Harrassowitz: 3-21.


(eds), *Temple Building and Temple Cult. Architecture and Cultic Paraphernalia of Temples in the Levant (2.-1. Mill. BCE)*, Abhandlungen des Deutschen Palästina-Vereins 41, Wiesbaden, Harrassowitz: 55-78.

