INTRODUCTION
1. SUMERIAN DIVINE EPITHETS
2. THE DATABASE RECORD
3. THE TEAM
4. BIBLIOGRAPHY
1. SUMERIAN DIVINE EPITHETS
The Mesopotamian deities reflect the entire natural world of Mesopotamia as well as the elements of Mesopotamian society and culture (Sallaberger 2003–2005: 296–299; Krebernik 2012: 46–48; Sallaberger 2020: 405–407). Their names alone, however, usually do not reveal their manifold powers and qualities. Specific linguistic determinations, the epithets, are needed in order to convey the full range of the deities’ manifestations and aspects.
Linguistically, epithets are adjectives and nouns used in appositions to a name (Gondos 1994; for relative clause see below). However, since a divine epithet can also replace the deity’s name, and since both may have the same structure (e.g., noun + adjective), the question arises of how to distinguish divine names from epithets. For instance, the divine name Ninmah (nin mah = “exalted lady”), was originally an epithet of the mother goddess used in the Old Sumerian city of Lagas (ca. 2500–2300), where her name was Ninhursaĝa (Selz 1995: 256). In order to distinguish epithets from names, the scribes’ classification principle is often, but not alway helpful. When the semantic classifier for divine names occurs before an expression, this can be understood as the name of a god. The problem here is that a deity can be referred to by different names in the same text. This is due to cases of syncretism or the differentiation of former epithets which become independent divine names. Examples are the interchangeable use of the names of the mother goddess Ninmah, Ninhursaĝa, Aruru or the names of the warrior god Ninurta, Ningirsu and Utaulu in the Old Babylonian period.
Formally, Sumerian epithets can be described according to different structures: adjective; noun; participle; noun(genitive); noun/participle + adjective; noun + noun(genitive); noun1 (+ noun2 (+...)) + participle and many, many more.
The greatest difficulty in identifying an epithet lies in the segmentation of sequences of epithets, especially with regard to attributive participle phrases. In this cases, the problem arises of how to define the semantic and syntactic border between epithets. For example, in Ninurta’s epithet saĝ-kal usu maḫ tuku, saĝ-kal “the foremost one” may act as the head of the expression usu maḫ tuku, which is formally a participle phrase meaning “having superior strength”. According to this interpretation, we are facing one single epithet (“the foremost one having superior strength”). However, it is grammatically possible to consider saĝ-kal and usu maḫ tuku as two distinct epithets of Ninurta: “Ninurta, the foremost one, the one who has superior strength”.
When a noun precedes a participle phrase, it is here usually considered as the head of clause. Consequently, the whole expression has been recorded as one single epithet in the database. This applies also when the head is followed by two participles whose meanings are particularly related, for example “warrior who pillages the cities and subjugates the mountains”.
While Mesopotamian divine epithets may appear in a standardised form (e.g., “gracious woman” as a standard epithet of Babu and Ninmarki), they usually do not develop into the standard double names (divine name + epithet) so typical of Greek deities (e.g., Zeus Keraunios “Zeus of the thunderbolt”; Schwabl 1978; Gladigow 1981: 1226–1229; Parker 2017: 28–31). Sumerian divine epithets, which could be compared with the Greek double names, are geographical epithets – particularly attested for Innana, the goddess of sex and war (e.g., Innana of Uruk; Ceccarelli 2021: 134).
Beside epithets in the actual sense, there are the hypotactic descriptions of a deity’s actions which occur as Sumerian nominalised clauses. These are collected in the database because they are appositions to a preceding noun, which, like participles, can express essential activities or qualities of a deity (Seux 1967: 16 for Akkadian relative clauses as epithets).
Semantically and conceptually, epithets are shaped by elements derived from various sources: typical aspects of human life or Mesopotamian society (e.g., “father” or “vizier of the god x”), general concepts of the deities (e.g., “shining”) or a specific aspect of a deity (e.g., Ninurta’s epithet “warrior”); mythical narrative materials, which can be concretised in various texts, can be a source of epithets. While some epithets are typical for a deity and can potentially occur in every text as almost fixed lexical items (‘standard epithets’), others are chosen according to the specific context and purpose of the text (‘contextual or situational epithets’).More generic epithets are the ‘honorific epithets’ (e.g., “lord”).
Despite the lack of a general theoretical framework regarding the function and meaning of Mesopotamian divine epithets, however, we can benefit from remarkable theoretical contributions from Classical and Religious Studies. Especially in recent years, interest in divine epithets has increased in these disciplines (Belayche et al. 2005; Wallensten 2008; Parker 2003; ib. 2017; Bonnet et al. 2018; Bonnet et al. 2019). Divine epithets have a 1) semantic, 2) pragmatic, 3) internarrative and 4) stylistic function.
1) Generally, epithets specify both ‘standard’ and less common qualities as well as local manifestations of a deity (Gladigow 1981; Parker 2003; ib. 2017: 9–17). Furthermore, Gladigow (1981) illustrates how the attribution of the same epithets to different deities establishes categorial cross-connections between them.
2) Epithets have a striking meaning within the cult (Gladigow 1981; Parker 2003; ib. 2017). In invocations, epithets enhance a goal-oriented communication with a deity by referring to the needed divine aspects and functional relations between the deity, its cult place or its deeds and the supplicant (‘contextual epithets’; Gladigow 1981: 1226–1229; Budde 2011). For instance, the epithet “merciful” is supposed to awaken the mercy of the deity.
3) Internarrative function. Divine ‘internarrative epithets’ explicitly refer to a deity’s deeds and a mythical narrative material. I analyse these epithets applying the hermeneutical approach of Annette Zgoll und Christian Zgoll and the categories developed in Christian Zgoll’s study of mythical narratives. In his theory (Zgoll 2019), a narrative can be analysed as a sequence of smallest action-bearing units, which he calls hylems, consisting of the basic structure logic subject + logic predicate (+ logic object), e.g., “god x does y”. Special hylems, called hyper-hylems, evoke or summarise a whole narrative material (Zgoll 2019: 186–197), e.g., the sentence “The god Ea created mankind”, attested in some incantations, is a hyper-hylem summarising the whole myth creation of mankind by Enki/Ea, which is concretised in different texts (e.g., Enki and Ninmaḫ, Atram-ḫasīs). The meaning of such epithets goes beyond their semantics and becomes relevant in shaping an effective communication between humans and deities by addressing the particular divine aspect and power needed in the context of the incantation. This finding is in line with the pragmatic function of epithets in cult and invocations, as Gladigow (1981) already outlined.
4) Epithets as stylistic device in Greek poetry have been studied by Parry (1928). Stylistically, Mesopotamian epithets are used to anticipate the divine name in climax and repetitions.
2. THE DATABASE RECORD
The database collects information related to the semantic, pragmatic and internarrative functions of the epithets. Generally, each epithet appearing in a text is recorded in an own entry. There are some exceptions: 1) epithets appearing in repeated passages are recorded in one entry, and the parallel lines are reported in the field ‘note’. 2) The honorific epithet lugal-a-ni "his (= of the ruler) lord” in royal inscriptions has not been recorded.
The record is structured in four sections:
A number of fields, which may not be immediately obvious, are clarified below.
2.1. DEITY
This section gives information on the deity to whom the epithet refers.
- ‘Divine type’: This concerns the general character of the deity, e.g., ‘sun god’, ‘weather god’, ‘mother goddess’. It allows to search the epithets of deities belonging to a specific divine typology and/or deities which are syncretised with each other. If you want to search for the epithets of the ‘mother goddess’ you can chose this divine type instead of searching for each single deity belonging to this divine type (Ninmah, Aruru, Ninhursaga etc.). Divine types in capital, e.g., INNANA, refers to deities who were usually syncretised with Innana or considered her particular manifestations.
2.2. EPITHET
This section contains information related to the semantics and the divine aspect of the epithet. It also gives information on relations with other deities, the king, and to particular categories of places (e.g., temple name, city name).
‘Epithet’: This is the standard form of the epithet without morphemes related to its particular synctatic function in the clause. In the case that the epithet is restored with a degree of uncertainty, an asterisk is affixed. Emesal words are ‘translated’ in standard Sumerian.
‘Form in context’: This is form with morphemes as it occurs in the text.
‘Semantic field’: This field refers to the basic semantics of the epithets. In case of multi-word epithets (like noun+adjective or head+participle phrases), the value refers to the head of the epithets, i.e., the first word of the epithet. For instance, the semantics of the epithet gud ḫuš furious bull” is ‘animal’.
The semantic categories here are based on the classification elaborated for the WordNet® Lexical Database for English of the University of Princenton because these categories are already used by numerous lexical databases for a variety of languages (Global WordNet Association).
There are, however, differences in the semantic categories used. First of all, the basic semantic categories are those related to nouns, not to verbs as in WordNet, because nouns are by far the majority of epithets. Nouns also appear very often as the syntactic head of participle phrases used as epithets. While WordNet syntactic categories ‘noun’ and ‘verb’ are adopted here, the syntactic categories ‘adjective’ and ‘adverb’ are not; adjectives and adverbs are recorded in the closest noun or verb semantic categories.‘Specific semantic field’: In this field, the semantic categories are specified in order to allow more precise searches. This allows, for instance, to search not only for ‘animal’ epithets, but also for ‘cattle’, or to search not only for the semantic category ‘person’, but also for specific professions.
‘Divine aspect’: This is the divine aspect expressed by the epithet. In the case of multi-word epithets, the divine aspect is usually defined according to the semantic field of the adjective or participle describing the epithet’s head. For instance, the divien aspect of the epithet gud ḫuš furious bull” is ‘wrath’.
‘Language’: The default value for this field is standard Sumerian (emegir). If an epithet contains emesal words, the value is ‘Emesal’.
‘Syntactic function’: Values in this field are:
- Byname = this is used for divine name used as epithet, e.g., ‘Nunamnir’
- Epithet = default value
- Epithet (2 verbs)/(3 verbs) = this value indicates that the epithet consists of two/three participle clauses which are synctactally and/or semantically strictly related to each other
- Non-verbal predicate + copula = this refers to descriptions in form of a non-verbal predicate followed by the copula (e.g., an-za₃-še₃ maḫ-am₃ "Great up to the farthest reaches of the heavens”). Despite not beeing eptithets in a strinct sense, such expressions express the same information of the epithets. Moreover, they may occur without copula and/or within a sequence of epithets.
Leaving this field empty extends the search to all entries.
2.3. GRAMMAR
This section provides information related to grammatic and syntactic aspects of the epithets.
- ‘Structure’ and ‘Participle’: Abbreviations used in this fields:
- Abl = ablative
- Adj = adjective
- Adv = adverb
- Attr. = Attributive
- AttNP = attributive noun phrase
- AttPP = attributive participle phrase
- Cop = copula
- Divine N = divine name
- B = verbal base
- B(m) = marû verbal base
- B.B = reduplicated ḫamṭu verbal base
- Dir = directive
- Equa = equative
- Gen = genitive
- Loc = locative
- N = name
- N2 = name 2
- n.p. = non-person class
- Obj = object (absolutiv)
- p = person; person-class
- Pl = plural
- PartP = participle phrase
- p.p. = plural
- Part = participle
‘Adj.’: Adjectives are only recorded in the structure when they appear alone or in simple structures like ‘N + adj’. When a noun(phrase) is part of a more complex structure, adjectives are not recorded. For instance, there are structures like ‘N + N(gen) or ’N(head) (+N2) + part.’, but not ‘N + N + adj.(gen)’ or ‘N + adj.(head) (+N2) + part.’. This choice has been made to avoid the multiplication of standard structures.
‘N(head) (+N2) + part.’: Many epithets are participle phrases. In this structure, the head of the participle phrase is also the agens of the participle. For instance, am a₂ ḫuš il₂-il₂ ‟Wild bull raising (its) fierce horns”, where am = N(head), a₂ ḫuš = N2, il₂-il₂ = part.
‘N(head) + N(agens)-3.n-p. + part.’: In this structure, N(head) is the head of the participle phrase, but the agens is N(agens). The possessive relation of the agens with the head is expressed by the resumptive pronoun. For instance, u₄ me-lem₄-bi nir ŋal₂ ‟Storm whose awe-inspiring sheen is majestic”, where u₄ = N(head) [= head], me-lem₄-bi = N(agans)-3n-p. [agens+resumptive pronoun], nir = N3, ŋal₂ = part.
‘N(agens)-3.p. (+ N) + part.’: In this structure, the first noun ist the agens of the participle phrase. It is followed by a third person resumptive pronoun referring to the external head in the main clause (other pronouns are also possible). For instance, enim zi-da-ni {d}suen-ra sa₆ ‟Whose reliable words pleases Suen”, where enim zi-da-ni = N(agens)-3.p., {d}suen-ra = N, sa₆ = part.
‘N(obj.) (+ N2 + ...) + part.’: In this structure, the first noun of the clause is the object of the participle. It also describes participle of so-called compund verbs (N = nominal element as absolutive). For instance, ḫi-li du₈-du₈-a ‟Adorned with attractiveness”, where ḫi-li = N(obj.), du₈-du₈-a = part.
‘N(x-case.) (+ N2 + ...) + part.’: In this structure, the first noun of the clause is neither the ergative nor the absolutive. For instance, {d}en-lil₂-gen₇ dim₂-ma ‟Created like Enlil”, where {d}en-lil₂-gen₇ = N(x-case), dim₂-ma = part.
‘N + AttNP(N)’: In this structure, an attributive noun phrase modifies the preceding noun(phrase) in a way similar to English constructions like ‟car park”. This structure can be translated with a prepositional phrase. For instance, am a₂ gur-gur ‟Wild bull with knobby horns”, wher am₂ = N, gur-gur = AttNP(N). The AttNP itself can have different structures.
‘N + AttPP(N(head) (+ …) + part.)’: In this structure, an attributive participle phrase modifies the preceding noun(phrase). For instance, en me sag₇ nu-di ‟Lord with divine powers which cannot be scattered”, where en = N, me = N(head of th participle phrase), sag₇ = (…), nu-di = part.
‘N + N(gen - attr.): This is formally a genitive construction, but the genitive can be understand as an attribute of the preceding noun and can be translated as a prepositional phrase with the preposition ‟with”. For instance, en {ŋeš-tu₉}ŋeštu daŋal-la₂ ‟Lord with broad wisdom”, where en = N, {ŋeš-tu₉}ŋeštu daŋal-la₂ = N(gen - attr.).
‘Bahuvrihi (N + N(gen))’: This (and variant structures) describes a bahuvrihi compound. A bahuvrihi denotes the referent by specifying its characteristic or quality. The default translation adopted is ‟the one with…/who…” For instance ŋidru nam-lugal-la "The one with the sceptre of kingship”, where ŋidru = N, nam-lugal-la = N(gen).
‘Divine name’: This and related categories refer to divine names used as epithets and allow to easily retrieve them.
‘lu₂’: Epithets beginning with lu₂ have separate categories in order to easily retrieve them.
- ‘Position’: This specifies whether the epithet appears directly before or after the divine name (in the same line)
Some of the ‘Stuctures’:
- Variations:
2.4. COMPOSITION
This section gives information on various elements related to the actual composition and to the speicfic passage where the epithet occurs.
‘Point of view’: This specifies how the epithet is referred to a deity.
- 1.p. = the epithet occurs in a first-person speech of a deity.
- 2.p. = the deity is addressed directly in a second-person speech (e.g., appeal, dialogue, praise).
- 3.p. = the deity is not addressed directly (e.g., narratives, direct speeches about a third deity).
‘Internarrative’: This field shows whether the epithet refers to a known narrative material.
‘Reference’: This field specifies the underlining narrative material.
‘Sequence of epithets’: This field shows whether the epithet occurs in a sequence of more epithets. Sequences of epthets may extend over one or several lines. If just one or two lines without epithets occur within a long sequence, the sequence is considered as uninterrupted.
‘Sequence: This field specifies the actual lines of the sequence.
3. THE TEAM
Dr. Manuel Ceccarelli (PI).
Taichuan Tang BA (Student assistant).
4. BIBLIOGRAPHY
Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Commission. Neither the European Union nor the granting authority can be held responsible for them.