Hermeneutic Techniques

By Enrique Jiménez | Make a correction or suggestion

How to cite
Jiménez, E., 2013, “Hermeneutic Techniques,” Cuneiform Commentaries Project (E. Frahm, E. Jiménez, M. Frazer, and K. Wagensonner), 2013–2024; accessed June 24, 2024, at https://ccp.yale.edu/introduction/hermeneutic-techniques. DOI: 10079/6wwpzvj
© Cuneiform Commentaries Project (Citation Guidelines)

This page provides an overview of the hermeneutic techniques used in Mesopotamian commentaries. The various and heterogeneous ways in which Mesopotamian commentators explained words or phrases are divided here into two categories: “philological explanations” and “non-philological explanations.” Both categories are discussed below and illustrated with examples taken from sundry commentaries. Following this exposition is a discussion of the different goals of the commentaries.

The means chosen to explain a word in Babylonian commentaries depend first on the particular problem posed by the word in question (the explanandum), secondly on the possibilities offered to the commentator by the lexical tradition, and thirdly on the new meaning that the commentator wishes to propose. Individual commentaries can use different hermeneutic techniques and often include both literal and non-literal explanations.

The examples provided on this page have mostly been excerpted from editions prepared for the Cuneiform Commentaries Project. To access the full text, please click on the CCP number that appears after each quotation.


Philological Explanations


Spellings and writings

On a basic level, commentaries address the issues posed by the complexity of the cuneiform script, fulfilling the same function as glosses in other texts. The simplest way of explaining a difficult spelling is to provide an unambiguous spelling of the same word: in the following example an archaizing spelling (ku-us-ZA-at) is explained by means of a more recent univocal writing:


(7) ku-us-sà(ZA)-at : ku-us-sa-at
(7) ku-us-sà-at (= Šumma Izbu XIV unknown) (is the same as) ku-us-sa-at


On occasions, commentaries render syllabically a word written with a logogram. Sometimes, these glosses explain unconventional or ambiguous logograms, as in the following case, in which the logogram nunuz is said to have one of its several possible meanings, that of “son”:


(21) [mār(nunuz) šarri(lugal)] kussâ(⸢gišgu.za) ul i-ṣab-bat : nunuz : ma-ru
(21) [In “The descendant (nunuz) of the king] will not seize the throne" (= Šumma Izbu XIV unknown), nunuz means “son.”
(CCP 3.6.3.D l. 21)


In the case of lines written in Sumerian, these basic explanations may include one or several Akkadian translations of the line. Although the degree of accuracy of the translation(s) may vary depending on the scope of the commentary, the texts occasionally lay out the word matches between the Sumerian original and the Akkadian translation, as a way of demonstrating the translation’s rigorousness. Thus in the following example the Sumerian sentence si é-gar₈-bi til-la is translated into Akkadian, and then the translation is justified by matching each of the words in the original with those of the translation:


(18) si é-gar₈-bi til-la : el-lam-mu-u (19) šá nu-ú-ru la-ni-šu ú-qat-ta-a : si : nu-ú-rù : é-gar₈ : la-a-nu (20) bi : šu-u : til : qa-tu-u
(18) si é-gar₈-bi til-la (means) “Elammû (an epithet of Sîn), (19) whose figure brings the light to an end,” (i.e., “it is eclipsed”) (since) si means “light,” egar means “figure,” (20) bi means “his,” and til means “to bring to an end.”
(CCP 4.2.A.a ll. 18-20)



At the semantic level, the easiest way of explaining an uncommon word is to provide a synonym for it: examples of explanations of lemmata through synonymy abound in Babylonian commentaries. In the example below, the verb uptanarrad, “(the snake) constantly causes fear,” is explained by means of the near synonym uptanallaḫ, “(the snake) constantly terrifies.”


(42)up⸣-ta-nar-rad : up-ta-na-al-làḫ
(42) “(the snake) constantly causes fear” (= Šumma Ālu XXIII 57) means “it constantly terrifies”
(CCP 3.5.22.A l. 42)


In cases of words that had become obsolete, commentaries may provide their contemporary equivalents. This is the case of several entries of the commentary on the Code of Ḫammurapi, which pair some words of the Code fallen out of use with their first millennium equivalents:


(r 3’) [... ra-b]a-an-nu : ḫa-za-nu : ⸢x⸣ [...] (…) (r 5’) [... nu-ma]-at : ú-na-a-tú : [...]
(r 3’) [...] rabannu (= Hammurapi Code §23) means “mayor” [...], (…) (r 5’) [...] numātu means “possessions”; [...]
(CCP 5.1 r 3’ and 5’)


Sometimes commentaries give several alternative renderings of one and the same logogram, or several synonyms of one and the same word. These explanations can occasionally take the form of long chains of equations, in which the tertium comparationis is either their similar meaning or, frequently, their Sumerian equivalent. Thus the following passage explains the word nešelpû, “slithered,” first by citing its infinitive (nešalpû, “to slither”), and then by providing two synonyms, “to cross” and “to go.” These two words are seamlessly followed by a third one, “to slide,” which is intruced by the commentator after justifying that both “to slith” and “to slide” have the same Sumerian equivalent, gir₅-gir₅:


(23) né-šel-pa-a (24) né-šal-pú-u : e-te-qu : né-šal-pú-u : a-la-ka : gir₅-gir₅ : na-ḫal-ṣu-u (25) gir₅-gir₅ : na-šal-pú-u
(23) “Slithered” (nešelpû) stems from “to slith” (nešalpû), (24) which means “to cross;” “to slither” (nešalpû) (also means) “to go.” (Moreover), gir₅-gir₅, which means “to slide” (nahalṣû), (25) can also mean “to slith” (našalpû).
(CCP 4.2.A.a l. 23-25)


Commentarial glosses may address simultaneously orthographic and semantic issues. Thus in the following example a logogram is first rendered into Akkadian, and then the commentator specifies which of the several meanings of the Akkadian word is intended:


(10) si : e-še-ri šá a-la-ku
(10) si means “to go straight,” said of walking
(CCP 4.2.A.a ll. 10)


Occasionally words are explained not by their synonyms but by words from the same semantic field, either to elucidate a word or to open the path to further speculation. These words may even include antonyms. Thus in the next example the adverb “no” is explained as “yes”:


(30) ul-la : an-na
(30) “No” means “yes.”
(CCP 4.2.A.a l. 30)


Morphological Matters

At the morphological level, inflected nouns and conjugated verbs may be explained by means of their citation forms, i.e., their infinitives or nominatives. Occasionally the preposition ana (muḫḫi) is used to introduce the citation form (see Technical Terms and Signs). Thus in the following example the verb irimmu, “(he who) is merciful,” is said to derive from the infinitive rêmu:


(15) i-ri-im-mu : ana re-e-mu
(15) “(he who) is merciful” (irimmu) (= Theodicy 41) derives from “to be merciful” (rêmu).
(CCP 1.4 l. 15)


Occasionally commentaries parse words in a way that does not entirely conform to the standards of modern linguistics. For instance, they claim that obscure words derive from common nouns, even if that derivation is grammatically impossible. In the following example the word kuppu, “spring,” from Theodicy 23 is said to be related to the similarly sounding, but etymologically unrelated, kappu, “bank,” which is then equated with nāru, “river”:


(7) ku-up-pi : ana kap-pa : na-a-ri!
(7) “spring” (kuppu) (= Theodicy 23) (is related to) “bank” (kappu), which means “river” (nāru).
(CCP 1.4 l. 7)


Homonyms, Etymology, and Etymography

Homonyms or similar sounding words are frequently equated in commentaries, sometimes regardless of their actual meanings. Occasionally near homonyms do have similar meanings, as in the following two examples:


(o 3) i-ḫar-ru-ur : i-gar-r[u-ur]
(o 3) “It is convulsed” (iḫarrur) (= Sagig X 2) means “it writhes” (igarrur)



(r 35) un-nu-tu : un-nu-šú
(r 35) “To be faint” (unnutu) (= Šumma Ālu XXV unknown) means “to weaken” (unnušu).
(CCP 3.5.25 r 35)


Closely related to the homonymic explanations are the commentary entries in which a word is artificially etymologized by dividing it into several parts and explaining the different parts individually. That is the case of the following passage, in which the word ḫurdatu, “vulva,” is explained by means of the similarly sounding phrase ḫurri dādi, “cavity of the darling”:


(42) šal-šiš ḫur-da-ti : ḫur-ri da-du (43) da-du : ma-ra
(42) Thirdly, “vulva” (ḫurdatu) can mean “cavity of the darling” (ḫurri dādi), (43) (where) “darling” means “son.”
(CCP 4.2.A.a l. 42-43)


Such (pseudo-)etymological explanations are similar to the technique known as notarikon in later rabbinic exegesis. Both are based on the belief that the smaller components of a word can reveal important information about its meaning (or that, as in the language of Borges’s story The Analytical Language of John Wilkins, “each word is defined by itself”). As opposed to the rabbis, however, Mesopotamian commentators counted on two languages for their speculations, Akkadian and Sumerian, whose connections, however tenuous, they fully explored in commentaries. The interplay between Akkadian and Sumerian provided the Mesopotamian commentators with a very large set of hermeneutical possibilities: in the following example, the rare word kisimmu (a Sumerian loanword into Akkadian), is translated as “sherpherd’s fold,” and then this translation is justified by explaining that “etymographically” kisimmu means “depth of the sheep”:


(3)ki-si-im | kisim₅ | ki-sim-mu : tar-ba-ṣu re-⸢ʾu⸣-[ú-ti] (4) aš-šum ki : šup-lu : si : im-[me-rù]
(3) kisim₅, read /kisim/, means “sour milk” (?) (= Ea VIII 162), which means “shepherd’s fold,” (4) on account of (the fact that from its components) ki means “depth” and si means “sh[eep].”
(CCP 6.1.41 ll. 3-4)


In the following example the word “at dusk” (barāri) in an omen is reinterpreted as “not at its time” by dividing it into two and by providing separate interpretations for both parts: the first one (bara) would mean “not,” the second (ri) would mean “time”:


(1) ¶ 30 iḫmiṭ(tab)-ma ba-ra-ri i-ta-aʾ-dar (2) antallê(an.gi₆) šar(lugal) akkadî(uriki) (3) ba-ra : la-a : ri : a-dan-nu (4) ina la a-dan-ni-šú ūmu(u₄) 12.kám ūmu(u₄) 13.kám antallû(an.gi₆) iššakkan(gar)-ma
(1) “If the moon is early and eclipses at dusk (barāri), it is an eclipse for the king of Akkad” (= Enūma Anu Enlil XV 1). bara (in Sumerian) means “no,” ri means “time”; (thus it means that) an eclipse takes place not at its time, i.e., on the 12th or 13th day.
(CCP 3.1.15 ll. 1-4)


In other passages the dissected word is not the explanandum, but rather its Sumerian equivalent. Thus in the following explanation, which comes from a commentary on an astrological text but is also attested in medical commentaries, the Akkadian word “diarrhoea” (nišḫu) is explained by means of a notarikon analysis performed on its Sumerian equivalent, šà-sur:


(18’) niš-ḫu : su-a-lu : ṣa-na-ḫu : šà.⸢sur⸣ [: niš]-ḫu (19’) šà : lib-bu : sur : ši-[te-en]-⸢nu
(18’) “Diarrhoea” means suālu-disease, i.e., “to excrete.” The signs šà.sur mean “diarrhoea,” since šà means “belly” and sur means “to urinate.”
(CCP 3.1.41 l. 18’-19’)


Occasionally the etymological speculations use forms of words that are unattested or grammatically aberrant, but that appear possible when considering the peculiarities of the cuneiform script. Thus in the following entry an analytic breakdown of the phrase “hail stones from the sky” (abnū tīk an-e) reveals that it is related to the act of giving birth. To arrive at this conclusion, the commentator uses a number of rather bold hermeneutical moves, such as pairing the word for “sky” (whose logographic writing is an-e) with the word “seed,” since the word for “seed” is equated in an obscure lexical list with the rare word ammu, “people,” which could theoretically – but never actually – be written an, i.e., the same sign used to write “heaven.”


(46) ab-nu ti-ik-ku šamê(an-e) : ab-nu (47) ab : ba-nu-u šá a-la-du : nu : ze-rù : u-gu₄ : ti-ik-ku : u : ba-nu-u (48) gu : a-la-du : an : an-e : an : ze-ri
(46) (In) “Hail stones from the sky” (abnū tīk an-e), “stone” (abnu) (47) (can be explained as) ab, i.e., “to create” said of giving birth, and nu, i.e., seed. (Likewise,) in u.gú, which means “rain” (tīku), u stands for “to create” (48) and gu stands for “to give birth.” an means “sky” (šamê), (but) an (also) means “seed.”
(CCP 4.2.A.a l. 46-48)



Non-Philological Explanations

The preferred method of definition in modern dictionaries and encyclopedias, the paraphrase, features in Mesopotamian commentaries as well. The most common technical term to introduce paraphrases appears to be the relative pronoun ša (see Technical Terms and Signs). Clauses introduced by ša offer a definition of the explanandum which is not restrained by the limits of philology or by the lexical equations of the word being explained. Thus in the following example the omen “a snake lays an egg” is said to refer to a snake “which gives birth in a man’s house”:


(38) pelâ(nunuz*) iš-tak-na : (39)šáina bīt(é) amēli() ul-la-du
(38) “It lays an egg” refers (39) to (the snake) which gives birth in the man’s house.
(CCP 3.5.22.A ll. 38-39)


Paraphrases are frequently used to redefine the scope or meaning of the original text. In the following example, a commentary on an omen regarding a female dog giving birth “to one” is redefined to mean “to one female”:


(o 5)kalbatu(nig) ina bīt(é) amēli() 1 ulid(ù.tu) bītu(é) šū(bi) (o 6) [ṣu-mi-ra-tu]-šú ikaššad(kur-ad) : šá ištēt(1*-et) sinništa(munus) ul-la-du
(o 5) “If a bitch gives birth to one in a man’s house, that house (o 6) will achieve its desires” (= Šumma Ālu XLVIII unknown) refers (to the bitch) that gives birth to one female.
(CCP 3.5.48 o 5-6)


As in the previous example, paraphrases often seek to specify something that in the base text could have several conceivable interpretations. Thus the commentary on Marduk’s Address to the Demons interprets the line from the base text (anāku asalluḫi) ša ina nāri ubbabu kēna u ragga, “(I am Asalluḫi), who in the river purifies the righteous one and the wicked” (CCP 2.2.1.A rev 14) as “it is said on account of the (river) ordeal” (aššu(mu) ḫur-sa-an iq-ta-bi). The interpretation in these cases seems often to be based on the commentator’s imagination and ingenuity.

Paraphrases play an important but still understudied role in Mesopotamian hermeneutics: a close study of them may provide valuable perspectives on what A. L. Oppenheim called “the impossibility of gauging adequately the conscious and subconscious associations inherent in the words of a dead language.”1 In this respect, the figurative interpretations which sometimes feature in Babylonian commentaries are particularly revealing. These figurative interpretations, i.e., associations between two seemingly unrelated concepts whose tertium comparationis is not of linguistic nature, are relatively rare in cuneiform commentaries. One occurs in the commentary to Enūma eliš and in related texts, where the “wind” with which Marduk carries away Tiamat’s blood is said to represent a certain “race,” in all likelihood because of the connotations of “speed” perceived in both concepts:


(23) [ú-par-ri]-iʾ-ma uš-lat dāmī(múdmeš)-šá šá-a-ru il-ta-a-[nu] ana bu-⸢us-ra⸣-a-ti uš-ta-bil ([š]a-ru aššum(m[u])) li-is-mu šá ina kislīmi(itigan) ūmi(u₄) 4.kám x [o o o o o o o o o]
(23) In “He severed her arteries | And made the north wind carry (her blood as symbol of) the good news” (= Enūma eliš IV 131-132), (“the wind” represents) the race that [takes place] on the 4th of Kislīmu […].
(CCP 1.1.A l. 23 [K.4657+ o 23-24 // BM 66606+ o 5-9])



Goals of the Interpretations

What is the purpose of the Mesopotamian commentaries? Many of them seem to be concerned exclusively with philological matters. This is the case, for instance, of the commentary on Ludlul bēl nēmeqi (CCP 1.3), which aims at ensuring the correct interpretation of a difficult text. Other commentaries, while dealing extensively with philological issues, also attempt to provide new layers of interpretation: thus the commentary on Enūma eliš I-VII, which contains many glosses on difficult words and expressions, also features sophisticated astronomical and theological speculations, such as correlations between mythological events and contemporary ritual actions.

On the other hand, the main focus of many commentaries seems to be to demonstrate the consistency and pertinence of the texts commented upon, rather than to gloss difficult words: this is the case of the commentary on Enūma eliš VII. The seventh tablet of Enūma eliš consists of a series of names of Marduk followed by short descriptions of the qualities of the god in these various hypostases. The commentary seeks to demonstrate that every single word of the descriptions is in fact contained within the name in question. Thus the 15th name of Marduk, Tutu-ziku, is described in the epic as “the god of sweet breath, the lord who listens and consents” (dTutu-ziku (…) il šāri ṭābi bēl tašmê u magāri). The commentary then equates the syllables of the name with every word of the description:


dingir i-lu dingir (i.e., the determinative) (means) “god”
tutu₁₅ šá-a-ri tu (means) “breath”
dudu₁₀ ṭa-a-bu du means “sweet”
dingir be-lu dingir (i.e., the determinative) (means) “lord”
zi še-mu-ú zi means “to listen”
zi ma-ga-rù zi means “to consent”
(CCP 1.1.B ii 14-19)


The same goal, namely to demonstrate the internal consistency of the base text, lies behind some highly sophisticated commentaries with a very clear agenda, such as CCP 4.2.A, cited several times throughout this page. That text seeks to prove that every word and ritual object mentioned in an incantation to facilitate childbirth is in fact related to the act of giving birth. Many commentaries on divination texts share this concern to demonstrate the validity of their base texts. To do so they set out to prove that the protasis and the apodosis of the omens are related, i.e., that one can be inferred from the other. It is a well-known fact that many omens do display a clear connection between protasis and apodosis, which is based on factors such as homonymy or symbolic association: for instance, an extispicy omen states that two holes (pilšū) on the gall bladder represent the omen of Apišalîm, “which Narām-Sîn defeated by menas of (digging) a hole (in its wall)” (YOS 10 24 9). The Mesopotamians were aware of these kinds of syntagmatic relationships between protases and apodoses in omens bequeathed to them by tradition, and commentaries often struggle to discover connections in those cases in which the rationale for the sequence protasis-apodosis is not so obvious. The elucidation of this relationship can take very sophisticated forms, which may include sophisticated astronomical or ritual interpretations, or quotations from literary texts in which a term from the protasis and a term from the apodosis co-occur. The following example displays one such oblique line of reasoning:


(13) libbī(šà) libbī(šà) ištanassi(gù-gù-s[i) qāt(šu) eṭemmi(gidim) (ø)] (14) [(ø) š]á-né-e ištar(d15) imât(gam) libbu(šà) : ištar(d15) : is-suk mul-mul iḫ⸣-te-pi ka-r[as-sa]
(13) In “He cries once and again “my belly, my belly!” – [hand of a ghost], (14) emissary of Ištar. He will die” (= Sagig IV 10). “Belly” (in the protasis is related to) “Ištar” (in the apodosis), (as in) “He shot an arrow that pierced her belly” (quotation from Enūma eliš IV 11).
(CCP 4.1.4.B o 13-14)


The passage seeks to demonstrate that the apodosis can be inferred from the protasis. To do so, it states that a word from the protasis (libbu, “belly”) is related to a word from the apodosis (Ištar), since they both appear together, mutatis mutandis, in a line from the Epic of Creation. The line in question describes how Marduk pierces Tiāmat’s “belly” by means of an “arrow”: the latter word, written mul-mul in the commentary, is reinterpreted as symbolizing the astral aspect of the goddess Ištar.2


These types of commentaries represent some sort of “secondary elaboration” – to borrow Evans-Pritchard’s expression – of their base texts: even if an omen’s prediction fails, the error would never cast a doubt on the validity of the omen or of the divination system. On the contrary, commentaries prove that the divination system has firm logical foundations. In the system described by Evans-Pritchard failures in the predictions reinforce rather than undermine the belief in the system, since the failures can be shown to conform to the system by means of a “secondary elaboration,” i.e., an attribution of the errors to external factors. Mesopotamian commentaries produce “secondary elaborations” for instance when dealing with omens with impossible predictions or circumstances, such as those that speak of the movement of fixed stars of constellations: commentaries in these cases equate constellations with planets, thus making sense of omens otherwise absurd. Other commentaries provide keys for an eventual reinterpretation of an omen, thus protecting its prediction beforehand from failure. By furnishing the text with various layers of interpretation, Mesopotamian commentators created a system of “converging causal sequences,” which in theory would allow to explain any possible failure of the prediction. The richness of the cuneiform script supports these reinterpretations to a large extent: in the example below, the month to which the omen applies (the fifth month, Abu, written itine) is said to mean “that month,” since the sign ne is equated with the Akkadian word “that” in a number of lexical lists: 3


(49) u₄ 22.kam u₄ 14.kam u₄ 25.kam u₄ 15.kam ina a-re-e qa-bi (50) abu(itine) arḫu(iti) šu-a-tune šú-a-tuina ṣa-a-ti qa-bi
(49) “The 22nd day (equals) the 14th day; the 25th day (equals) the 15th day,” it is said in a mathematical table. (50) “The month of Abu” (written itine) (means) “that month,” (since) “ne (means) “that,” it is said in a bilingual list.
(CCP 3.2.1.A ll. 49-50)


Guide to Further Reading

For a detailed overview of the hermeneutic techniques used in Mesopotamian commentaries, see Frahm, 2011E. Frahm, Babylonian and Assyrian Text Commentaries. Origins of Interpretation. Ugarit-Verlag, 2011.: 59-85. Earlier but still useful discussions include Cavigneaux, 1987A. Cavigneaux, Aux sources du Midrash: L'herméneutique babylonienne, Aula Orientalis, vol. 5, pp. 243-255, 1987. and George, 1991bA. R. George, Babylonian Texts from the folios of Sidney Smith. Part Two: Prognostic and Diagnostic Omens, Tablet I, Revue d'Assyriologie, vol. 85, pp. 137-167, 1991.. Important studies on individual technical terms are Gabbay, 2012U. Gabbay, Akkadian Commentaries from Ancient Mesopotamia and Their Relation to Early Hebrew Exegesis, Dead Sea Discoveries, vol. 19, pp. 267-312, 2012. and Gabbay, 2014U. Gabbay, Actual Sense and Scriptural Intention: Literal Meaning and Its Terminology in Akkadian and Hebrew Commentaries, in Encounters by the Rivers of Babylon: Scholarly Conversations between Jews, Iranians, and Babylonians, U. Gabbay and Secunda, S. , Eds. Mohr Siebeck, 2014, pp. 335-370..

  • 1. A. L. Oppenheim, Mantic Dreams in the Ancient Near East, in The Dream and Human Societies, G. E. von Grunebaum and Caillois, R. , Eds. University of California Press, 1966, pp. 341-350. P. 345.
  • 2. However the exact way in which the goddess’s planet, Venus, is related to the “Pleiades” (mulmul) is not specified in the commentary.
  • 3. Compare also CCP 3.1.47 l. 43’, which reinterprets the month of Kislīmu (itigan) as “that month,” since the sign gan, read as kam, means “that” in Sumerian.