# Defining the Hittite "Pantheon", its Hierarchy and Circles: Methodological Perspectives

Livio Warbinek1

**Abstract**: For the Hittite religion of the "Thousand Gods of Ḫatti" the scholarship has identified different ways of categorization: State pantheon, Local cults, "circle" and numeric group are the most widely used categories based on several criteria, such as linguistics, geography, and cultural milieu. The present paper aims to better define the state of the question about the hierarchy within the Hittite pantheon on the one hand, and to further investigate the notion of "circle" in the Hittite religion on the other, whose analysis has raised some questions and has led to different interpretations.

## 1. Hittite religion

In the field of Anatolian religion of the 2nd millennium BC, the Hittites created a peculiar religious structure: the "Thousand Gods of Ḫatti"2 . Indeed, scholarship provides different interpretations of the *religio hethitica* in Anatolia according to different criteria, as recently highlighted by Cammarosano (2021, 94) «The gods […] can be considered from different perspectives, depending on whether the focus is on their typology, geographical areas of attestation, or cultural milieu». Contextually, different ways to describe this religious system have emerged: the most common concepts3 employed in literature are those of State pantheon, Local Cults, numeric groups, and circles.

According to Gilan (2019, 179) «by 'Hittite religion' we mainly refer to the religious practices that are attested in the state archives of the Hittite capital, mostly relating to the king and the royal family or to 'state religion'. This is mostly due to the nature of the Hittite textual evidence. No private archives and very few administrative texts have been found so far in Hittite Anatolia», whereas for Taracha (2009, 86) it was a «peculiar product of political theology, an amalgam of Anatolian, Hurrian, Syrian and Mesopotamian religious traditions»4 . It should not be forgotten that this religious amalgam had been possible and successful thanks, above all, to the rich cultural contacts of the


<sup>3</sup> Taracha 2009, 38 *passim*; Hutter 2013, 183 "State Pantheon", 187 "state cult"; Rutherford 2020, 185; Hutter 2021, 29.

<sup>4</sup> See also Hutter 2003, 115; Taracha 2010, 858.

Livio Warbinek, University of Verona, Italy, livio.warbinek@univr.it, 0000-0002-5472-9177 Referee List (DOI 10.36253/fup\_referee\_list)

FUP Best Practice in Scholarly Publishing (DOI 10.36253/fup\_best\_practice)

Livio Warbinek, *Defining the Hittite "Pantheon", its Hierarchy and Circles: Methodological Perspectives*, © Author(s), CC BY 4.0, DOI 10.36253/979-12-215-0109-4.16, in Livio Warbinek, Federico Giusfredi (edited by), *Theonyms, Panthea and Syncretisms in Hittite Anatolia and Northern Syria. Proceedings of the TeAI Workshop Held in Verona, March 25-26, 2022*, pp. 181-191, 2023, published by Firenze University Press, ISBN 979-12-215-0109-4, DOI 10.36253/979-12-215-0109-4

Hittite kingdom with other civilizations, from the Aegean Sea to the Euphrates and from the Black Sea to the Orontes, during several historical events.

Within this wider context, the creative process of the Thousand Gods seems to evolved according several possibilities for interaction among cultures5 developing a complex religious system which necessitates further consideration.

# 2. Divine hierarchy?

The epigraphic sources we have – as well as the iconographic ones, when available – clearly highlight a precise order of importance among deities6 . Particularly, I am referring to both the sequence of divine offerings in the cultural texts, which «was never accidental and it reflected the divine hierarchy» (Taracha 2009, 39), and the Hittite state treaties, which «list the witnessing gods in an order determined by certain set principles» (Popko 1995, 90). According to Popko's analysis, already at the time of the reign of King Arnuwanda I and Queen Ašmunikal, the canon of the pantheon was close to that which appears in the later treaties, thus pinpointing the creation of the official pantheon to the Middle Hittite period7 .

Stage by stage, all the gods were apparently organized according to their order of importance8 in a structure of the divine world which seems to have the fixed shape of a pyramid, with only minor variations: at the top were the main gods, from the Stormgod to his wife and their extended family; below, we encounter the second-rank gods, such as the Mother-goddesses, the Solar deities; then gradually all the others from the War-god to the god of vegetation, the spirits, the tutelary figures, and the evil gods; finally, at the bottom the ancient departed kings9 . Not less importantly, the «Hittites did not generally refer to 'kingship' among the gods» even though «the Storm-god is occasionally addressed as 'King (of Heaven)' and his spouse as 'Queen' (e.g. KUB 6, 45 i 10-12)» (Beckman 2004, 313b). This scheme can also be applied to the local panthea, where in a «Hittite town, you tend to find a simpler structure, with a top-tier, comprising a storm god and either a sun deity or a mother goddess; a 'tutelary deity' or deities, […] sometimes a god of war, and various local deities associated with mountains, rivers, and/or springs» (Rutherford 2020, 185-86). This hierarchical vision is apparently derived from the conception of the divine among the Hittites, because the «inhabitants of Anatolia imagined the world of the gods in likeness to the real world around them»10 and the «kinship and the hierarchy in the pantheon structure closely resembled actual social relations» (Taracha 2009, 80).

However, even if this representation could be valid in general, it portrays a palimpsest that returns a synchronic and flat picture of what was, in all likelihood, the result of a diachronic process of stratification. As is the case with the different features of Stormgods, tutelary deities, and so on, the gods should not – in my opinion – be grouped together axiomatically, because if the available evidence offers a sufficiently clear picture, this picture highlights as many differences as there are similarities. As correctly point-

<sup>5</sup> See Schwemer 2008, 147-48 and *Introduction* in the present volume.

<sup>6</sup> Haas 1994, 633; Taracha 2009, 39.

<sup>7</sup> Popko 1995, 90. See also Taracha 2009, 83-3; 86; Taracha 2010, 861; and Devecchi 2015, 48-9 for an outline of the *evocatio* of the divine witnesses.

<sup>8</sup> As already expressed by Gurney 1977, 4-5; Lebrun 1980, 50-1; Archi 1993, 7.

<sup>9</sup> See Popko 1995, 90-1; 112; Taracha 2009, 32; 38; Mouton 2014, 19; 27.

<sup>10</sup> Taracha (2009, 80) with reference to KUB 13.4 I 21-22.

ed out by Beckman (2004, 313a) «No single hierarchy prevailed among the gods in all circumstances». If it is true that the hierarchical order of the divine witnesses in the treaties of the Empire period was standardized11, this does not allow us to consider it valid everywhere during and throughout the history of the Hittite kingdom. At present, the division of Hittite history into different periods and several local cults does not allow us to conclusively establish a hierarchical vision of all aspects of the Hittite religion: it is perhaps better to accept the definition of convenience of "Thousand Gods of Ḫatti" without speculating further on, apparently, uneven paths.

## 3. Panthea and languages

When studying the *religio hethitica*, a linguistic subdivision within the Thousand Gods of Ḫatti may be attempted, according to the different putative "ethnic groups" of the Hittite kingdom: Hattians, Hittites, Palaeans, Luwians and Hurrians. What we want to stress is that classifications based on linguistic criteria have always existed in literature12, even among scholars who do not declare it directly. Klinger (1996), in his examination of the Hattian religious milieu, chose to focus, instead, on the socalled *Kultschichten* of the Hittite religion, thereby recognizing the inconsistency of the attempt to define "pantheon" as a list or group of theonyms only because those gods seem to bear names with a common origin. Not only is the cultural milieu too complex to be defined merely by a linguistic identity. Even when using language as a tool, there are several ways one can proceed, and «an important criterion for attributing a given deity to a specific ethnic tradition is the language of her cult» (Taracha 2010, 859), not of her name. Unfortunately, however, the language of cult is not always evident, because we cannot access all levels of religious practice within the Anatolian societies.

All in all, the use of languages and linguistic analysis for the study of religion and panthea is a powerful tool, but it requires some caveats:

a) Since "language is not ethnicity" (a statement already well-underlined by Hutter 2003, 211), it should not be used as the unique criterion to describe a culture or a group;

b) Linguistic analysis can be applied to one theonym, not to an entire ethnic group, and therefore to build up a whole divine group is quite a difficult operation to achieve;

c) It is quite difficult to assess linguistic diversity through the common Hittite sources at our disposal.

According to these points, we should not speak of a Hattian, Hurrian, Palaean, or Luwian pantheon for the Hittite kingdom because they were not panthea, but divine names in different languages gathered in a whole religious system whose name for Hittites was the Thousand Gods of Ḫatti. For instance, Taracha (2009, 107) convincingly argued that «there was no one pantheon shared by all the Luwians, only individual deities worshiped in all of the Luwian territory». A Hurrian pantheon probably existed in the Hurrian lands, but what was imported and adopted in Ḫattuša cannot be addressed as a pantheon. More controversial is in my opinion the case of the possible Palaean pantheon: the corpus of Palaic texts specifically describes the festival for Zi-

<sup>11</sup> See Gurney 1977, 4-6; Beckman 2004, 313a.

<sup>12</sup> Laroche's (1946/47) subdivision into gods of different origins hides linguistic criteria. See also Gurney 1977, 7-16; van Gessel 1998, (Part 1) X; Hutter 2003, 218-20; Beckman 2004, 311a-b; Taracha 2009, 107-08.

parwa and the Palaean cult, but this is not enough to identify it as "pantheon" until we outline the Palaean religious system with certainty13.

This methodological issue regards, more generally, the ways to identify a pantheon. As already argued, the definitions found in literature are "glottocentric" even when given by authors who call for caution in this regard. The glottic criterion is quite difficult to set aside. For example, even Hutter's highly authoritative attempt at listing Luwian gods (2003, 219-20) cannot avoid using language as at least one of the criteria: together with gods that have generically Anatolian or opaque names (Pirwa, Šanta) and gods whose names have a clearly different etymology (Zilipura, Kamrušepa), the list features virtually all deities whose name can be analyzed as Luwian, which makes the Luwian etymology a "sufficient condition" for the definition.

Far for claiming that any linguistic categorization should we avoided, it is important to stress that the limitations of this approach, as well as its inevitability, should be always kept in mind. Beside recognizing the linguistic origin of a name, any study of a pantheon should always be aware that these do not always help identifying the geographical areas in which a cult existed, and, where possible, include both the criteria of cult language in which a deity was addressed14, and the textual contexts and traditions in which the deities are grouped and worshiped.

## 4. Divine groups

Finally, gods can be collected by geographical criteria and source typology, as well as in homogeneous or heterogeneous sets.

The former type of approach is based on the analysis of the cult inventories, which help us recognize the so-called Local Cults15, regarding «certain deities in relation to one or more specific towns» (Cammarosano 2021, 5). Since the official Hittite religion was a huge melting pot of multi-ethnic beliefs, the concrete aspects of the cult were in all likelihood performed locally, and in the descriptions of local cults «you tend to find a simpler structure, with a top-tier, comprising a storm god and either a sun deity or a mother goddess; a 'tutelary deity' or deities, […] sometimes a god of war, and various local deities associated with mountains, rivers, and/or springs» (Rutherford 2020, 185-86).

The second type of approach consists in collecting deities in homogeneous or heterogeneous groups. Homogeneous groups include deities with the same typology16, or gathered according to numeric groups17, even though these classifications do not match any Hittite structure18. Criteria for homogeneity vary and may include also the linguistic one19, but homogeneity itself may be more apparent than real, as local beliefs


<sup>13</sup> For an outline of the beliefs of the Palaeans see Archi 1993, 5-6; Taracha 2009, 58-9.

<sup>14</sup> Goetze 1953, 263; Archi 2004, 11; Taracha 2009, 36.

in Anatolia «must have been surely heterogeneous, reflecting the ethnic differentiation of the population of the land of Ḫatti» (Taracha 2009, 50). In the context of heterogeneous groups of gods, on the other hand, it is common to find the term "circle" in the scholarship, frequently used for a group of deities often linked to a town or a prominent god; often, the gods in these circles are quite different from each other and share only the presence in a given text or group of texts20.

# 4.1. "Circles"

In the religious system of the Hittite Kingdom, there are three main "circles" to deal with.

The first is that related to the god Ḫuwaššanna21. In the centre of Ḫupišna, identified as the classical Kybestra and modern Ereğli (East of Konya)22, the main deity Ḫuwaššanna/gaz.ba.a23 was surrounded by the following gods24: dutu, du, dlamma, <sup>ḫ</sup>ur.sagŠarlaimi, zababa, Lallariya, Awatta, Kupilla, Ašdutta, Ḫarduppi, Tunapi, Muli, Imralli, Lilaya, Waša(l)iya, <sup>ḫ</sup>ur.sagŠarpa, *hantezziuš* dingirmeš (i.e., the primeval gods Anna, Aruna, Zarnizza, ídŠarmamma).

The prominence of a god or a town is also a common element of the second "circle". This putative circle has been defined in several ways according to the prominence of the deity Pirwa25 or of the town Kaniš/Neša26, while the Hittite texts refer specifically to the "Gods of Kaniš" dingirmeš(-*aš*) uru*Kaniš*27. According to different Hittite texts, this group28 includes: Pirwa, dmunus.lugal (Ḫaššušara), Kamrušepa, Ašgašepa29, Maliya30, Šiwat, Šuwaliat, Ḫašammili, Išpant (Išpanzašepa), d 7.7.bi, Ḫalki, Ilaliya, Tarawa. However, the number of deities grows if we also take into account the gods from the lists and the theophoric names of the Cappadocian tablets in the Old Assyrian Colony period31.


Finally, "circle" can refer to the deities belonging to the *kaluti*-lists32 of the Hurrian Storm-god, his spouse, and their local *parhedroi.* In these Hurrian lists we can collect the following deities for Teššub: Tašmišu/Šuwaliyat (duraš), Kumarbi, Šauška, Sun-God (Šimige), Moon-god (Kušuḫ), dnin.gal, Ea, War-god (Aštabi), Ninatta-Kulitta, bulls, twelve gods of Netherworld; and for Ḫebat: Nabarbi, Šaluš-*Bitinḫi*, Damkina, Umbu-Nikkal, Ḫudena-Ḫudellara, Allatu (Lelwani), Aya, Išḫara, Ištar, Šauška, Šuwala, Ibrimuša, Tiyabenti, and her daughters Allanzu and Kunzišalli33. Apparently, the *kaluti* do not seem to have any proper geographical location, as the other so-called circles of Kaniš and Ḫupišna do. However, this is not really the case because the order of the processions in Yazılıkaya «corresponds to the order of a standard list of gods (socalled kaluti)» (Taracha 2009, 94)34. But the divine reliefs at Yazılıkaya did not represent the Hittite State cult; it rather follows the Hurrian dynastic pantheon35, which in turn is probably connected Hurrian deities of the *kaluti*-lists.

The definition of "circle" is obviously quite vague. The "circle of Pirwa" as well as the "circle of Ḫuwaššanna" put too much emphasis on the role of a single deity, based on limited and circumstantial evidence. The references to Kaniš or Ḫupišna are quite suitable, but positively generic even though they originated in local contexts 36. Furthermore, even if we assume that the gods of the putative circle did belong together on a functional level, most of them are also quoted in other texts outside those of the circle37, so it would be risky to presume that they always had an original connection with the geographical areas of interest (Warbinek 2022, 13).

Originally, the concept of "circle" was a definition of convenience – first used by Laroche (1946/47, 67: "cercle de Pirwa") and then gradually adopted by several scholars in different works38.

The circle of Kaneš is certainly the one that received most attention, but others labels have been employed to describe it, such as "Kanishite lists" (Goetze 1953, 264), "Kanesite pantheon/gods" (Popko 1995, 55; 88; Taracha 2010, 859-60); "Kreis der Götter von Kaniš" (Klinger 1996, 157); "Pantheon von Kanish" (Otten 1971, 32); "Gottheiten/Pantheon von Kaniš" (Haas 1994, 281; 412; 613). Perhaps, the gods of Kaneš formed a pantheon during the Old Assyrian Colony period when each Anato-


<sup>32</sup> Hurrian Teššub and Ḫebat, see Trémouille 1997, 201-10; Wegner 2002 and Taracha 2009, 102. See also Haas (1994, 347-49) "Der Götterkreis der Ša(w)oška von Ninive" and (1994, 402) "Kreis des Wettergottes (von) Manuz(z)i".

<sup>33</sup> Laroche 1948; Gurney 1977, 17-8; Haas 1994, 332-33; Trémouille 1997, 105-07; 202-03 with note 713; Taracha 2009, 118-19.

lian town had its own cult39, but they just became a Local Cult with the integration of Kaniš into the State religion of Ḫatti.40 Haas himself alternated the definition of "Pantheon von Kaniš" with that of "Gruppe der Gottheiten von Kaniš" (1994, 439, 613). In my opinion, the Hittite definition "Gods of Kaniš" dingirmeš (-*aš*) uru*Kaniš* should be preferred to any other label, which is also the choice made by Cammarosano (2021, 84).

The situation with other alleged circles is different. We don't know much about the origin of the one of Ḫuwaššanna, which can be more easily described as a local cult41. Hutter himself, who also used Ḫuwaššanna's circle definition (2003, 244), later placed (2013, 186) "Ḫuwaššanna's circle" in quotation marks and then *de facto* avoided its use in his most recent works42.

Finally, it is important to stress that «le *kaluti* ne signifie pas une liste abstraite et n'a rien à voir avec les litanies. Le *kaluti* était réellement, c'est-à-dir matériellement, l'ensemble des divinités, objets et lieux sacrés connexes à une divinité précise» (Trémouille 1997, 205); and this is justified by what looks like a sort of conclusion43 at the end of the libations:

KUB 32.92(+) (// KUB 32.85 + KBo 20.52 obv. III 6'/7' – *CTH* 664)44 rev. 5'-6':

5' [*ka-a-š*] *ša* dingirmeš lúmeš *ka-lu-*⸢*ti*⸣[-*iš*] 6' *qa-ti*

[Diese] *kaluti*-Liste der männlichen Götter ist zu Ende. (Wegner 1995, 100)

Admittedly, the technical term45 *kaluti* means "circle, closed group, cohort, community, round of offerings" (HEG K, 471-72; HED K, 33-5)46, with its denominal verb *kalutiya*- "to lump together (for worship), to celebrate as a group, to treat jointly (for cultic purposes), to make the rounds of " (HED K 471-72; HED K, 33-5; IEED 5, 130; CLL, 99)47.

KBo 15.59 (*CTH* 628) III? 7'-8': 7' íd*Al-da ka-lu-ut-ta* lugal [-*uš*]

8' 1 ninda.gur₄.ra *par-ši-ya* …

«The king breaks a loaf 〈to〉 the *circle* of the river Alda»48


However, it must be acknowledged that *kaluti* can also be translated as "line", "list", or "set" in accordance to the «principle of naming the gods (Teššub's circle) and goddesses (Ḫebat's circle) in the order of importance, from the greatest to the minor gods» (Taracha 2009, 118), as we can see from Wegner's translation of *kaluti* here:


und für die gesamte Reihe des Wettergottes, auch Bier libieren sie vor gesamten Reihe. (Wegner 2002, 273-75)

We hence seem to be back to the concept of divine hierarchy. According to Trémouille (1997, 201) «par *kaluti* on entend la liste de toutes les divinités et entités liées à une divinité plus importante» and it corresponds to Hittite *arḫa*-. However, the term *arḫa*- "Grenze, Gebiet" (HEG I, 55-6), "Line, Boundary" (IEED 5, 245), and its cognates *irḫatt(i)*- "row, series, circuit, in a row, by turns"; *arḫai*-/*irḫai*- "to go down the line, to make rounds, to list, to treat in succession" (HED I, 130-1; IEED 5, 245); and Luwian *irḫatta*- "circle" (CLL 99) are never used to state or portray a divine group (as for *kaluti*, above KUB 32.92(+)). In addition, *kaluti* and *arḫa*- are not equivalent according to HED (K, 34) and Kammenuber (1996, 47). However, *kaluti* appears once in relation with the infinitive *irḫawanzi*:

IBoT 1.2 (*CTH* 684)50 III 10-13:


The king 〈proceeds〉 to treat three with offerings in sequence: the Maraššanda River, the Tutelary Deity of the River, and (their) circle. (McMahon 1991, 193)

Already Goetze (1953, 274) translated the present *kaluti* as "circle", followed here by McMahon, but this is even more interesting given the presence of the rarely-attested infinitive *irḫawanzi*, which usually occurs in this textual evidence as lugal-*uš irḫawanzi tiyazi* "the king proceeds in order to treat with offerings" and for which the verb *tiyazi* «was certainly intended» (McMahon 1991, 196).

So, are we dealing with a line, a row, a procession, or rather a proper "circle"? In my opinion, this text can be a strong piece of evidence for equating *kaluti* and *arḫa*- as "circle" when we deal with textual evidence related to Hurrian gods. Therefore, at the present stage of research, the so-called *kaluti* divine groups are the only ones that may be safely defined as "circles"*.* However, Haas (1994, 389) with his definition of «Das Gefolge oder der Kreis der Ḫebat» attempted to render both possible concepts, which is yet another indication of the importance of terminological caution when dealing with a complex object of study such as the Hittite religion. Clarity on the criteria used for classification and on their limits, on the other hand, is and remains of paramount importance.

<sup>49</sup> Wegner 2002, 272-77.

<sup>50</sup> Goetze 1953, 274; McMahon 1991, 192-96.

#### 5. Conclusions

In the Anatolian pantheon of the Thousand Gods of Ḫatti, theonyms with Hittite, Luwian, Hattian, and Hurrian etymology can be found. Different etymologies are often present in names that belong to alleged groups, and, for this reason, the classification of a deity as Hittite, Luwian, Hattian or Hurrian is quite complex, even ambiguous, and possibly misleading51. Similarly, we cannot speak of Hattian, Hurrian, or Luwian pantheon in terms of different panthea because they were not panthea in the classical way, but rather collections of divine names and figures gathered in a whole religious system whose name was the Thousand Gods of Ḫatti, or in specific subsets of said system.

Hierarchies existed, but a generalized hierarchical model would be simplistic and simplifying: even if a divine hierarchy could be recognized in the tradition of particular local centres (Taracha 2009, 38), we should avoid a single pyramidal reconstruction for the *religio hethitica* of the 2nd millennium BC.

As for other types of groupings, although circles have been described in literature, for both homogeneous and heterogeneous groups of deities, the definition should be used only for specific homogenous ones. Indeed, the gods belonging to the so-called circles of Kaneš or Ḫupišna cannot be regarded as such: they are also mentioned in other texts, taking part in different events, showing different features at different ages and in different locations.

#### Bibliography


<sup>51</sup> A point already well stated by Cammarosano (2021, 63-4) «A god who was originally rooted e.g. in the Hattian culture underwent transformations along his life through the Hittite centuries […] both 'from below' and 'from above'».


Trémouille, Marie-Claude. 2002. "Una cérémonie pour Huwaššanna à Kuliwišna." In *Silva Anatolica. Anatolian Studies Presented to Maciej Popko on the Occasion of His 65th Birthday*, ed. Piotr Taracha, 351-69. Warsaw: Agade.

van Gessel, Ben H.L. 1998. *Onomasticon of the Hittite Pantheon 1-2*. Leiden-New York-Köln: Brill.

van den Hout, Theo P.J. 1995. *Der Ulmitešub-Vertrag. Eine prosopographische Untersuchung*. Wiesbaden: Harrassowitz (StBoT 38).

Warbinek, Livio. 2022. "The -*šepa* theonyms in the Hittite pantheon." *Vicino Oriente* 26: 1-19.

Warbinek, Livio, and Giusfredi, Federico. In press. "Maliya, Malja, Malis, Athena. From Kizzuwatna to the Aegean: Borrowings, Translations, or Syncretisms?" *Asia Anteriore Antica*.


Wegner, Ilse. 2004. *Hurritische Opferlisten aus hethitischen Festbeschreibungen, Teil III: Das Glossar*. Roma: Multigrafica editrice (ChS I/3-3).

