Research Showcases Divination

4.1. If a pig carries a reed and enters a man’s house… Observations on some structuring devices in Babylonian omen lists by Nicla de Zorzi

4.2. Stealing the Property of the Gods. Observations on a Top-Middle-Base Omen Sequence from an Old Babylonian Liver Model by Lucrezia Menicatti

4.3. Aspects of Creativity in the Assyrian Dream Book by Matthew Ong

4.4. If the Thunder Cries Like an Animal. Horizontal Connections and Vertical Arrangement in EAE 4: 17-20 by Lucrezia Menicatti

4.5. ‘From White, to Red, to Dark’ Colors and Their Interpretation in Manzāzu 2: 42-45 by Lucrezia Menicatti

4.6. Repetition as an Argumentative Technique in the Diagnostic Handbook by Eric Schmidtchen

4.1. If a pig carries a reed and enters a man’s house …

Observations on some structuring devices in Babylonian omen lists
Written by Nicla De Zorzi

How to cite:
De Zorzi, N. 2020, “If a pig carries a reed and enters a man’s house …Observations on some structuring devices in Babylonian omen lists,” Project REPAC (ERC Grant no. 803060), 2019-2024 at DOI: 10.25365/phaidra.230 (accessed day/month/year).

In second and first millennium BCE Mesopotamia, the texts most commonly associated with the practice of divination are omen lists written in Babylonian. These texts are a still largely unexplored source of information on ancient Mesopotamian scribal creativity and on scribal strategies of meaning and knowledge production.

Omens were formulated as conditional clauses whose protases or antecedents (A) describe a sign, and whose apodoses or consequents (C) give the pertinent prediction. Omen clauses draw on schematised sets of potential phenomena and match these as ominous signifiers with an equally selected set of signified predictions. The correspondence between sign and prediction is based on a likeness of some kind between them, on the semantic, phonemic or graphic level.

The scribes’ creativity was not limited to establishing connections between individual signs and individual predictions. In omen lists, this horizontal or syntagmatic level of omen production was interconnected with the vertical, paradigmatic axis, and we must still address major gaps in our knowledge here, since the sequencing of interdependent and partly repetitive omens has hitherto been studied only selectively, with a particular focus on sources from the Middle Bronze Age (Winitzer 2017). REPAC will offer a much-needed investigation of omen-organisation and omen-sequencing on the basis of Iron Age sources (WP 4).

One of REPAC’s main innovations is its focus on the micro-structure of Ancient Mesopotamian scholarly texts. In the following, a case-study based on a sequence of eleven interrelated omens dealing with the ominous behaviour of pigs will be presented. The omens are taken from Šumma izbu Tablet 22 (De Zorzi 2014). My aim is to demonstrate that the interplay of similarity and contrast between contiguous or near-contiguous textual elements is a significant operative principle in the process of text production in omen lists. Through a process of reverse engineering, a model for the development of this sequence over several stages can be proposed. These stages probably correspond to particular phases in the text’s history, as it gradually approached the stable form in which it was transmitted in the first millennium BCE:

22: 120) šumma šaḫû qanâ naši tibût nūnī u iṣṣūrī / iṣṣūrī nūnī / erbī nūnī ibbašši : tibûtu ibbašši
“If a pig carries a reed – there will be a swarming of fish and birds (var.: of locusts and fish); there will be a swarming (of animals)”
22: 121) šumma šaḫû qanâ našīma ana bīt amēli īrub bēlšu išarru
“If a pig carries a reed and enters a man’s house – (the house’s and the pig’s) owner will become rich.”
22: 122) šumma šaḫû qanâ našīma ištu bīt bēlišu ūṣi bēlšu mādūti ibissê / ibissâ immar
“If a pig carries a reed and exits from its owner’s house – its owner will suffer losses (var.: heavy losses).”
22: 123) šumma šaḫû rikis qanê našīma ana bīt bēlišu īrub bēlšu nēmelam immar
“If a pig carries a bundle of reeds and enters its owner’s house – its owner will make some profit.”
22: 124) šumma šaḫû rikis qanê našīma ištu bīt bēlišu ūṣi bēlšu ibissâ immar
“If a pig carries a bundle of reeds and exits its owner’s house – its owner will suffer losses.”
22: 125) šumma šaḫû ēri gišimmari naši meḫû itebbâm
“If a pig carries a palm frond – a storm will rise.”
22: 126) šumma šaḫû pitilta naši [su]nqu ina māti ibbašši
“If a pig carries a string plaited from date palm fibres – there will be a famine in the land.”
22: 127) šumma šaḫû kilibba našīma ina sūqi ittallak maḫīru ibbašši
“If a pig carries a large reed bundle and runs about in the street – there will be active trading.”
22: 128) šumma šaḫû kilibba našīma immelil tīb meḫê
“If a pig carries a large reed bundle and plays with it – there will be a storm.”
22: 129) šumma šaḫû kilibba našīma ištu bābi ana bīt bēlišu ūṣi bīt bēlišu išarru
“If a pig carries a large reed bundle and goes from the (city) gate towards its owner’s house – its owner’s household will become rich.”
22: 130) šumma šaḫû kilibba našīma ištu bīt bēlišu ana bābi ūṣi bīt bēlišu ibissâ immar
“If a pig carries a large reed bundle and goes from its owner’s house towards the (city) gate – its owner’s household will suffer losses.”

Let us look at single omens first: swarming birds, fish or locusts – we have different manuscript traditions for omen 120 – are a frequent prediction in omen lists. The reed-carrying pig may condition their appearance because of the habitat it shares with bird and fish at least – the reeds. “Swarming” is also conditioned by the semantic proximity between the words used for “to carry” (našû) and “to swarm” (tibûtu) – both involve the idea of “rising.”

The symmetrical pair 121 and 122 is based on 120 and add the dichotomy “entering” – “exiting:” in 121 the reed is being brought by the pig into the man’s house, while in 122 the reed is taken out of the house. The former suggests “getting rich,” the latter “impoverishment” (losses) – the underlying conventional analogy is intuitively intelligible. 123 und 124 are variations of 121 and 122. The “reed” is replaced by a “reed bundle,” and the dichotomy richness-losses is substituted by the dichotomy profit-losses.

The structure of 125 is that of 120: only we have a palm frond instead of a reed. The prediction refers to a topos also found in literature: date palms must face the wind and lose their fronds in a storm. The verb used, “rising” of the storm, itebbâm, is etymologically linked to the “swarming” of 120 (tibûtu). The structure of 126 follows that of 125: the string plaited from date palm fibres evokes a famine because palm fibres are eaten in extremis.

Omen 127 uses another word for “reed bundle” (kilibbu) and has the pig walk in the street. Street, sūqu, also means “market” in Akkadian, thus explaining the association with active trade. Omen 128 follows 127 structurally. The use of immelil “to play” in the protasis motivates the “storm” in the prediction: it is a topos in Babylonian to express the whirling of dust storms by this same word.

Finally, the couple 129 and 130, built around the binary opposition between the inward/outward movement of the pig, is a variation on 121-124: in this case, the “richness-losses” theme has a larger focus: it concerns not just the pig’s owner (121-124), but his whole household.

The following tables offer schematic presentations of this text’s structure with coloured graphs. Their purpose is to allow readers to follow the analysis more easily. In the first table, colours highlight the patterns of association between omens:

Let us now focus on the structure of the sequence. Starting with 120, the pig with the reed (120-122) begins a sequence of omens in which the pig carries a bundle of reeds (123-124), a palm frond (125), a string plaited from palm fibres (126), and a large reed bundle (127-130):

This vertical sequence is broken up at the beginning by the insertion of two binary pairs built around the dichotomy “entering” – “exiting” describing the pig’s movement: 121-122 and 123-124. These two pairs of omens are tightly interconnected: the underlying association is the same (“entering” – “richness/profit”, “exiting” – “losses”).

Omens 125-126 share the same structure of 120: the pig simply carries an item and no further action is taken into consideration. The tight connection between 120 and 125 becomes evident if one focuses on their respective apodoses: the apodosis of 125 repeats with a variation the “rising” (tebû) theme of 120.

Starting with omen 127, a differently structured vertical sequence is inserted (127-130). The element “If a pig carries a large reed bundle” does not form a protasis on its own as the two preceding elements do (125-126), but it is extended with different horizontal expansions describing various activities of the reed-carrying pig: running about in the street (127), playing with the reed bundle (128), going from the city gate towards the man’s house and the other way around (129-130).

The last couple of omens (129-130) are built around the binary opposition between the inward/outward movement of the pig, and, as already mentioned, are a variation on 121-124 with a larger focus (the household):

Let us now focus on the apodoses section of the sequence. On this level, there is significant repetition between the lines after the introductory omen (120): the dichotomy between richness-profit and loss is explored in 121-124, as well as in 129-130; both 125 and 128 deal with the rising of a storm and represent a variation on the “swarming”-theme of 120; 126-127 seem to introduce, with famine and active trading, a variant to the “richness/profit-losses” theme:

Importantly, the sequence generally avoids exact repetition. A theme is stated over and over, but with each appearance some aspect of it changes: the swarming of animals (120) turns into the rising of a storm (125 and, with variation, 128); richness (121) becomes first profit (123), then active trading (127); losses (122, 124) become famine (126). In the last couple of omens (129-130) the “richness-losses” theme has a larger focus: it concerns not just the pig’s owner (121-122; 123-124), but his whole household.

A closer look reveals that the apodoses create structures of thematic symmetry (ABB’A’B’’A’’B’’’):

Another pattern is inscribed within the sequence, as omens 121-125 and 128-130 can be shown to create a chiastic thematic structure: richness-losses (A) – rising of storm (B) – rising of a storm (B) – richness-losses (A). Interestingly, the two predictions inserted between the two components of this virtual chiasm, 126-127, are closely interconnected as they stand in a semantic opposition to each other: famine and active trade. In the highly repetitive sequence of the apodoses, they stand out, as their wording differs significantly from that displayed by the apodoses with which they share the general “richness-losses” theme (121-124, 129-130). If we now bring the protases back into the picture, we see that this opposition plays an important role in the structuring of the whole sequence: through this opposition the authors of these omens tie the second vertical sequence, 127-130, to the preceding one, 120-126:

The text achieves here (126-127) the linking of two separate vertical sequences by drawing on a binary opposition at the juncture. This technique, of which REPAC’s ongoing work is bringing to light numerous other cases, is revealing itself as an important structuring device in divinatory compendia.
A complex literary process is at work in this sequence: it involves textual expansion founded essentially on the use of similarity and analogy. There is a clear concentration in the vertical extension and the interconnections between the signs in the protases. This concentration on the level of the sign, however, implies a stronger schematisation of the predictions, which cannot keep up with the increasing complexity of the signs in the protases. The diviners’ hermeneutic code, their repertoire of analogical associations, was clearly not sufficiently elaborate. At the same time, our example demonstrates that Ancient Mesopotamian diviners did try to match the complexity of the protasis on the level of the apodosis by variation of single elements, and, particularly, through clever arrangement of relatively standard phrases.

In conclusion, we have seen how the authors of the omen sequence analysed here creatively operate with similarity and contrast between contiguous or near-contiguous textual elements, both on the horizontal, syntagmatic, and on the vertical, paradigmatic axis. Their modus operandi can be described as a creative process, as a construction of meaning principally based on analogical reasoning. Ancient Mesopotamian scribes conceived of this process as a matter of discovering pre-existent information written into the fabric of the world by the gods. In any case, this process, etically of creation, emically of discovery, is based on textual means and leads to knowledge of socially recognised validity.

Download here.


De Zorzi, N. 2014, La serie teratomantica Šumma izbu: testo, tradizione, orizzonti culturali (2 volumes). History of the Ancient Near East – Monographs 15, Padua.
Winitzer, A. 2017, Early Mesopotamian Divination Literature: its Organisational Framework and Generative and Paradigmatic Characteristics. Ancient Magic and Divination 12, Leiden/Boston.

4.2. Stealing the Property of the Gods

Observations on a Top-Middle-Base Omen Sequence from an Old Babylonian Liver Model
Written by Lucrezia Menicatti

How to cite: Menicatti L., 2021, “´Stealing the Property of the Gods´ – Observations on a Top-Middle-Base Omen Sequence from an Old Babylonian Liver Model, Version 01,” Project REPAC (ERC Grant no. 803060), 2019-2024, at DOI: 10.25365/phaidra.303 (accessed day/month/year)

Bu 89-4-26, CT 6 Pll. 1-3
© The Trustees of the British Museum

The Mesopotamian divinatory corpus includes a variety of different sources, among which the most significant are the so-called divinatory compendia. These are lists of omens structured as conditional clauses, in which the protasis describes the ominous sign and the apodosis gives the appropriate interpretation. The earliest extensive omen compendia were committed to writing during the Old Babylonian period (1900-1600 BCE). The majority of these texts were written on clay tablets of various dimensions, but clay models of the sheep’s organs inspected during the extispicy ritual have also been recovered, and these are occasionally inscribed with series of omens.
This is the case with Bu 89-4-26, 238, first published as CT 6 Pll.1-3 (Cuneiform Texts from Babylonian Tablets in the British Museum VI, London 1896 ff.). This is a well-preserved model of the sheep’s liver from the Old Babylonian period, excavated in Sippar (modern Tell Abu Habbah) and now stored in the British Museum. Its surface is divided into cases and inscribed with liver omens. Among these, three omens form a sequence concerning the part of the liver named naplaštum, ‘the View’, which corresponds to a vertical groove on the left lobe of the sheep’s liver and was the first zone inspected by the diviner performing the extispicy.
This sequence of omens, which we will discuss here, was edited in Nougayrol 1950: 29 (entries 11-13). The sequence corresponds to obv. 7 and rev. 8-9 in the edition published on In this sequence, the View is divided into three sections, and the same condition is observed first in the Top, second in the Middle, and third in the Base of the View. Similar sequences of entries arranged on the Top-Middle-Base scheme are widespread in extispicy texts (Winitzer 2017: 290-329). In the sequence under analysis, each protasis corresponds to an apodosis which includes two separate predictions.
The following study reveals the textual connections that build the structure of this sequence, on two levels. On the horizontal (or syntagmatic) level, I investigate the analogical connections between protasis and apodosis in a single omen, and I focus on the association of a given section of the View in the protasis with a certain figure in the apodosis. On the vertical (or paradigmatic) level, I consider the structure of this sequence as a whole and I focus on the vertical connections between the omens. This is meant to show the overall system of interpretation of the Top-Middle-Base paradigm in the sequence of apodoses, and the role played by repetition in the make-up of this passage. Furthermore, I will provide evidence for a particular use of the middle entry’s apodosis, which functions as a ‘pivot line’. As will be shown, this line anticipates part of the following apodosis and repeats part of the preceding one, thus playing the role of a structural medium between the two external elements in the sequence of predictions. In the following, the sequence is presented in a tabular form:

As the table shows, the protases describe a deep perforation (pališ-ma šutēbrû) in the different sections of the View (Top, Middle, and Base). A parallel sequence of apodoses corresponds to this sequence of protases. Each of the apodoses includes two predictions, the second one introduced by the conjunction šumma, and the subjects mentioned are figures related to the temple – a high priestess, the chief temple administrator and his wife, and a temple visitor. The first sequence of predictions forecasts repeated robberies in the temple, committed by the ēnu-priestess in the first entry, and by the chief temple administrator’s wife (aššat šagî) in the second and third entries. The first prediction adds the punishment for this crime, stating that the priestess will be captured and burned (iṣabbatūšī-ma iqallûši).
The Top of the View in the first entry of this sequence is equated with the high priestess. The Base of the Presence in the protasis is associated with the chief temple administrator’s wife in the apodosis of the second entry, while the third entry repeats the same prediction as the preceding one.
The second sequence of apodoses presents an interesting sequence of subjects as well. These predictions include illicit sexual activities between a high priestess and a second figure who is also related to the temple in some way. In the first entry, the subject is the chief temple administrator (šagûm) – and the second entry repeats the second one. In the final entry, it is ‘a temple visitor’ (muttallik bīt ilim) who ‘will repeatedly have intercourse’ (i-ta-na-ia-ak Cf. CAD N/2: 198, s.v. nâku 2, ‘to have illicit intercourse repeatedly’,) with an ēnu-priestess. Unlike the first sequence, none of the predictions in this sequence add the punishment for this illegal act.
The two sequences of apodoses are linked by their similar message, since the two predictions – ‘stealing the property of the gods’ and ‘having intercourse with an ēnu-priestess’ – both describe illegal acts that involve ‘stealing’ something that belongs to the gods. The ēnu-priestess was considered the human wife of the god she served, and thus belonged to the god in every respect. Her behaviour was extremely important; she was to act like a married woman, and her misconduct was an offense to the god himself (Sallaberger and Huber-Vulliet 2005: 626-627). The mention of this priestess being involved in illicit sexual acts is thus not unexpected in this sequence, since this prediction, just like the preceding one, implies a serious act of impiety. The parallelism of the two predictions is reinforced by the use of the same verbal stem (Gtn).
In summary, in the two sequences, the Top is equated with the high priestess and chief temple administrator, while the Base corresponds to the chief temple administrator’s wife and the temple visitor.

Interestingly, in the first sequence of apodoses, the second and third predictions are repeated, while in the second sequence, the first and second predictions are repeated.

The middle entry occupies a special position in the make-up of this passage. The first prediction in the apodosis (aššat šagîm asakkam ištanarriq) anticipates the following entry, which repeats this prediction verbatim, while the second prediction (šagûm ēnam ittanajjak) repeats the same prediction from the preceding entry.
The first entry is the only one including the phrase iṣabbatūši-ma iqallûši, ‘they will catch her and burn her’ in the first sequence of apodoses and predicts the death of the subject stealing from the temple (in this case, the ēnu-priestess). The death of the high priestess marks the beginning of this sequence and seems to represent the peak constituted by the sequence of predictions concerning robberies in the temple. The fact that this first entry includes elements that are not repeated in the following entries, namely the ēnu-priestess (who is also the most important figure in the sequence) and the prediction of her death, supports the focal role of this prediction in the first sequence of apodoses.
If we consider these two sequences of predictions as a coherent whole, we notice that the Top-Middle-Base scheme in the sequence of protases is interpreted as a figurative spatial scheme of closeness in the sequence of apodoses, in which the different subjects mentioned represent different steps in a scale of distance from the divine world. The first element of the first sequence and the final term of the second one function as two contrastive terms in this scale. The ēnu-priestess, who is the spouse of a god, represents the closest figure to the divine sphere in this hierarchy, while the ‘temple visitor’ represents the furthest one. The position of these two figures, at the beginning and at the end of the passage, functions as an additional element that reinforces their opposition. Also, the alliteration of velar /k/ and /q/ and of liquid /l/, together with the assonance of /a/, /u/ and /i/ in the final verb of the first apodosis in the first sequence (iQaLLuši) and the first word of the final apodosis in the second sequence (mutaLLiK) strengthens the link between the two predictions.

The two contrastive elements placed at the beginning and at the end of the respective sequences are interchanged with the repeated predictions aššat šagî(m) asakkam ištanarriq, ‘the chief temple administrator’s wife will repeatedly steal the property of the gods’, in the middle and third entries of the first sequence, and šagûm ēnam ittanajiak, ‘the chief temple administrator will repeatedly have intercourse with an ēnu-priestess’ in the first and middle entries of the second sequence. The middle entry contains both the repeated elements, of the first and of the second sequence, and thus builds a climax involving the two non-repeated elements. This line functions therefore as a ‘pivot’ line and plays the role of a structural medium between the two external contrastive elements of the two sequences, binding them together in a coherent passage. This results in two parallel sequences linked by the similarity of the image they express, namely, the idea of ‘stealing’ something that belongs to the gods.

Download here.

Selected Bibliography

Nougayrol, J. 1950, ‘Textes hepatoscopiques d’epoque ancienne conserves au Musee du Louvre (III)’ in Revue d’Assyriologie 44, 1-40.
Sallaberger, W./F. Huber-Vulliet, 2005, ‘Priester I. A. Mesopotamien’ in Reallexikon der Assyriologie und vorderasiatischen Archäologie Bd. 10, 617-640.
Winitzer, A. 2017, Early Mesopotamian Divination Literature. Its Organizational Framework and Generative and Paradigmatic Characteristics. Ancient Magic and Divination 12, Leiden/Boston.

4.3. Aspects of Creativity in the Assyrian Dream Book

Written by Matthew Ong
(UC Berkeley,

How to cite:
Ong, M. 2022, “Aspects of Creativity in the Assyrian Dream Book, Version 01,” Project REPAC (ERC Grant no. 803060), 2019-2024 at DOI: 10.25365/phaidra.338 (accessed day/month/year).

Abstract: This research showcase argues that the Assyrian Dream Book contains a number of interesting semantic properties reflecting what is known as ‘exploratory creativity’. This kind of creativity can be metaphorically understood as a person traversing a path across a space of conceptual possibilities. The features of the Dream Book which illustrate this creativity are found in omen protases and include processes of conceptual elaboration, abstraction, and contextual frame shifting. Via such processes, much of the Dream Book can be seen as an exploratory creative walk traversing numerous semantic domains linked by principles of creative association. The upshot of the research showcase is that Akkadian omen collections like the Dream Book crucially involve a heuristic exploratory element akin to what is found in creative storytelling.

1. Frame-shifting as exploring territory

The collection of dream omens known as the Assyrian Dream Book (Oppenheim 1956) is a good example of how elite Akkadian scribes engaged in the cognitive process known as exploratory creativity.
Exploratory creativity relies on the phenomenon of frame-shifting, where we reinterpret a given object or concept by viewing it against the background of a new contextualizing frame. A frame is a basic structure of cognition reflecting how humans actually interpret their world, namely in terms of integrated scenes or actions (Fillmore 1982). An example is the Restaurant frame, which involves a customer going to a restaurant, sitting down and ordering food, having the cook make the food, having the waitress bring it out, paying the bill, etc. All of the characteristic roles, actions, smells, emotions, and linguistic terms associated with this scene make up the Restaurant frame, and evoking any one of its characteristic elements evokes the entire frame.
Recontextualizing an object by viewing it as part of another frame can lead to a new set of associations linking the object to the new frame. A simple example would be looking at a tree and thinking of it first as a home for woodland animals. You think of the branches as places birds or squirrels can rest on, and the leaves provide shade for them under the sun. You might then switch to thinking of the tree as a source of timber. You think of the branches as things that can be cut down and turned into beams. The leaves then become annoying things that need to be thrown away. All throughout the frame-shift, the immediate object of your attention stays the same, but the conditioning frame under which it gets its broader meaning changes.
How should one think of exploratory creativity, and how does it relate to frame-shifting? Exploratory creativity is about creating something where you must repeatedly choose among multiple possibilities for how to proceed without either knowing in advance how those choices will play out in the end or having a fixed set of rules to make all your decisions for you. These ‘choices’ often depend on frame-shift. A useful metaphor for thinking of frame-shifting is an exploratory walk through unknown territory. When you reach the top of a hill that you initially couldn’t see beyond, you may decide to proceed in a new direction based on your updated perspective. You might have a habit of exploring the terrain every day by starting out on your walk in the same direction but finding yourself inevitably going on divergent routes due to minor random direction changes on the way. You may also find that on your walk, your decision of which way to go places more emphasis on near-term benefits and less on global ones (such as heading over to a nearby small hill with a shady tree for a slightly better vantage point rather than walking much longer under the hot sun in order to get to a tall mountain).
The above metaphor highlights how frame-shifting, like exploratory walking, is heuristic, stochastic, and opportunistic. It is heuristic in that to a large (but not total) extent, the mind does not make frame-shifts based on pre-set, abstract rules applicable across all contexts. Rather, it largely operates on immediate context that is sensitive to detail. It is stochastic in that the same person who considers the same general set of information can still, on a different occasion, decide to frame-shift in different ways. This has to do with the stochastic nature of many of our lower-order brain functions (Simonton 2003). Finally, frame-shifting is opportunistic in that it tends to be triggered by salient elements of the current mental scene or particularly strong associations involving part of that scene. Rather than considering the consequences of a frame-shift several steps down the line, the mind tends to go for what is immediately promising in terms of imaginative possibilities.

2. Example: wooden items to professional crafts

Frame-shifting is evident in a portion of the Assyrian Dream Book dealing with wooden items and professional crafts. The frame defining the omen protases starts out as Carpentry and switches to Craft/Profession based on the semantic categories of key lexical items in the protases. This process is illustrated below for the initial section of Tablet III (K.3941+4017: obv. i 1-18, cf. Oppenheim 1956: 263, 308), shown graphically in Figure 1. Note that the horizontal lines in the transliteration do not refer to rulings on the tablet but abstract divisions between frames.

Figure 1: Frame-shifting from Carpentry to Craft/Profession in the initial section of Tablet III (K.3941+4017: obv. i 1-18). Not all omen protases from those lines are depicted. Each frame is represented by a rectangle, where the elements of that frame are put within the rectangle. The area where the rectangles overlap contains elements common to both frames. One can imagine that the scribe composes the omens by mentally proceeding left to right across the category elements of the Carpentry frame (door, bed, etc.), and then doing the same for the Craft/Profession frame.

Figure 1 illustrates the two types of association between the Carpentry and Craft/Profession frames whereby the scribe composing the omens in lines 1-18 shifts his train of thought. The first association involves basic roles in the scene described by each frame. In the Carpentry frame, a carpenter creates wooden objects such as beds, doors, and stools. These wooden objects play the role of what the agent in the frame (the carpenter) physically produces. In the Crafts/Profession frame, a craftsman or skilled worker such as a leather-worker or seal-cutter exercises his trade skill to earn a living. In the new frame, these professions play the role of what general actions or work a skilled person performs to earn a living. The scribe starts out composing lines 1-7 thinking of various wooden objects, but eventually comes to see those objects as instances of what a carpenter produces (see the end of section 3 for more details). Viewing the carpenter as but one instance of a craft or profession evokes the Craft/Profession frame, giving the scribe a new category to enumerate. The second association between the frames is the grammatical construction ‘X + epēšu’ they both use to linguistically express what the agent in the scene does or produces. This shared grammatical construction helps highlight which elements in each frame are parallel. For instance, in the Carpentry frame X denotes an object built by a dreamer, while in the Craft/Profession frame X denotes the craftsman’s work which the dreamer performs.

One should note that the omen in line 8 dealing with the night watchman presents a complication to the above analysis, since it is not a prototypical craft or profession, nor is it the carpentry profession one might expect on the heels of omens about wooden objects. This may be an instance of path digression (see below), where the scribe follows a less obvious chain of associations for a short time before returning to his main train of thought. Alternatively, in the frame-shift the scribe may already have been thinking further ahead than we give him credit, where he wanted to organize the whole set of craft/profession omens according to some internal scheme. One possibility is that the scribe may have been assembling thematically related but pre-compiled sub-lists of omens, one containing lines 1-7 and the other 8-18. Frame-shift would have still motivated the scribe to concatenate the two lists even if the linear order of the latter one does not reflect what he might have written were he composing from scratch. Whether it was a case of impromptu thinking or pre-compiled lists, writing down the first list would have led the scribe to make a frame-shift and think of professions.
Elyze Zomer has suggested that the scribe may have aimed at parallelism between the two lists, as a door (1-2) is associated with the work of a night watchman (8), a stool (6) is associated with the work of a carpenter (15-16), and a boat (7) is associated with the work of a sailor (17-18). If this is so, the associative logic in the remaining omens is not clear. One also wonders why the scribe chose to produce a negative omen variant in 10-11 but not do the same for the remaining craft/profession omens.

3. Types of exploratory steps

In the Assyrian Dream Book the scribe uses a number of techniques for exploring a particular conceptual domain, including elaboration, abstraction, adjustment, and compression. The first three techniques can be illustrated with the minimally contrastive pairs in English given below. Sentence 1 is meant to be contrasted pairwise with each of 2-4:

1) John took his dog to the park and walked him there for an hour.
2) John took his dog to Central Park in New York and walked him there for an hour.
3) John took his dog to the park and walked him there (for an hour).
4) John took his dog to the beach and walked him there for an hour.

In elaboration, one adds perceptual detail to the scene. Thus in sentence 2 the added detail is marked in red. In abstraction, one renders the scene more schematic. Thus in sentence 3 the orange strikethrough refers to detail that was thrown out from sentence 1. In adjustment, one alters one parameter of a scene by switching the specific value but maintaining the type. Thus in sentence 4, the altered parameter is marked in green.
The fourth technique of compression takes elements from multiple scenes that have some analogous relation to each other and blends them into one thing. An example given by Fauconnier and Turner is the concept of dusk (Fauconnier and Turner 2002: 196). Humans get the concept of dusk by observing the analogical relation between the times when the sun goes down each individual day (e.g. the day before yesterday, yesterday, and today), and collapsing them into the idea of when the sun goes down during any day. Compression is different from abstraction in operating over multiple scenes in parallel rather than just one scene. While some information across the input scenes is no longer explicitly represented in the output, this is a result of constructing an ideal or archetype rather than reducing what we know about something specific.
We can find instances of elaboration, abstraction, and adjustment in Tablet III, K.3941+4017: rev. ii x+1 to x+7 (Oppenheim 1956: 264, 308). As with the example sentences, the colors below are to be interpreted by comparing the first omen pairwise against the succeeding ones:

The type of exploratory step taken in lines x+6-7 depends on what the rest of the omen is (which we do not have). Assuming there is no reference to taking a plow and seeding barley with it, talking about planting onions represents a simple switch in the type of crop cultivated.
One can observe compression at work in the previous set of omens dealing with wooden objects and professions. The frame-shift in that example from Carpenter to Craft/Profession is facilitated when the scribe compresses the specific wooden objects the dreaming man makes into a general wooden object, which allows the scribe to think of a carpenter engaging in his profession more generally.
Note that this means more generally, during the composition process it can take some time for the scribe to conceptualize what he is currently writing in terms of a fairly elaborate frame (such as Carpenter). True to the spirit of an exploratory walk, the scribe can find himself already having put down the first few omens in a new sequence on the basis of implicit association or semi-random thoughts before ‘realizing’ his new direction.

4. Path digression

In the Assyrian Dream Book, frame-shifts sometimes happen in unexpected ways. This can be likened to a digression from the main line of thought of the scribe. While the digression is often thought of as ‘accidental’ in nature, it is simply a frame-shift according to a conceptual axis other than the Gestalt organizing the current block (e.g. homophony, polysemy, homography). The example below comes from Oppenheim’s Tablet A, Sm. 2073 rev. y+10-14 (Oppenheim 1956: 273, 317). As before, the horizontal lines below refer to abstract frame divisions, not tablet rulings.

This block begins with a protasis about eating dust (SAḪAR) on its own, then moves to eating dust in a rubbish dump (SAḪAR tubkinni), and then eating leper scales. This last line is included in the block because the term for leper scales (saharšubbû), is based on the metaphor of falling dust. The line directly after returns to the main theme with a protasis about eating something which is semantically related to dust, namely sand (bāṣu).
One reason to think that in the mind of the scribe the omen about leper scales represents more of a temporary digression rather than permanent shift in conceptual direction is that he put a tablet ruling only after the omen about eating sand (cf. Oppenheim 1956: 317, 367). After that ruling, the theme changes to a much different topic, eating leather.

Figure 2: ‘If he eats the faeces of wild animals, he will have riches’. Greyed elements are frame elements that have been abstracted or suppressed. Roles in the main action of each scene (i.e. eating and having) are connected to scene participants via the assignment function r(verbal role) = scene participant.

5. Competing factors in exploratory creativity

The mind is regularly generating thoughts and images in a quasi-stochastic manner. Such thoughts are the source of exploratory creativity. They are governed by two basic factors and operate at two different levels, the individual and social (Table 1).

Suspension of judgment deals with our ability to suppress the pragmatic filter normally applied to our thoughts regarding purpose, acceptability, utility, etc. What is interesting deals with the fascination for certain thoughts based on factors like perceptual stimulation, social taboo, novelty, etc.
In the Assyrian Dream Book, the scribe schematizes the frame of an omen protasis to match it with a stock omen apodosis. Figure 2 shows how this works for the omen ‘If he eats the faeces of wild animals, he will have riches’ (DIŠ ŠE ú-ma-me GU7 NÍG.TUKU T[UKU-ši], Tablet A Sm. 2073: rev. y+27, cf. Oppenheim 1956: 317).
In general, the omen protasis, representing the content of a dream, can have wild content. But the content is schematized so that it can be connected to a stock apodosis via analogical correspondence. Figuring out how to do this is part of the extensive tradition of elite scribal hermeneutics. As a result of the abstraction and analogical mapping (here called signification), when thinking of a possible dream protasis for his omen the composing scribe is able to play off of suspension of judgment and what is interesting in a way not possible for other omen types or text genres. This is why dream omen composition can be considered exploratory creativity.

In Figure 2, the frame of Man Eating Animal Faeces is licensed because it is abstracted to Man Has Thing, which maps analogically to the frame Man Has Riches in the apodosis. This provides the necessary suspension of judgment. If a suitable analogical correspondence with the apodosis could not be found, the scene evoked by the protasis would be rejected as scandalous. Conversely, because the abstracted details are scandalous (what is interesting), the scene that evokes them is more likely to occur to the scribe relative to many other more mundane scenes that are also possible in dreams.
In particular, the initially paradoxical combination of negative protasis (eating faeces) with positive apodosis (acquiring riches) also seems to be motivated by suspension of judgement and what is interesting. In essence, combining scenes of opposite polarity does something ‘unexpected’ relative to the way omen hermeneutics often works, and it is allowed since the basic structural features of many of the stock omen apodoses (e.g. losing or acquiring something, change of state in mind or body) can be easily elaborated in both positive and negative ways (e.g. eating faeces versus eating good food, becoming sad versus becoming happy). Ann Guinan suggested something similar occurred in Šumma ālu, arguing that flipping the polarity in a single pair of corresponding elements of the protasis and apodosis was an easy way to ‘complicate’ the significance of the omen (Guinan 2014: 119 and Guinan 1990: 231)

6. Cognitive mechanism for generating the Assyrian Dream Book

Overall, we should reconceptualize how the Akkadian scribe constructed the Assyrian Dream Book, switching our model of what is going on in the scribe’s head and on the tablet from the left side of Figure 3 to the right side.
In the figure, the upper two rectangles reflect the mental processes the scribe goes through when compiling the omens. The lower two rectangles show the output on the tablet (which in both models is the same, consisting of a strict linear sequence of omens). Vertical arrows connecting the upper and lower rectangles illustrate implicit decisions the scribe makes as he shifts attention between written content on the tablet and his own internal thoughts. In particular, the new model specifies feedback processes where unanticipated frames can be evoked in the scribe’s mind midway through composition, and this in turn will a new set of omens on the tablet.
Using the analogy of sequential omen composition as walking across the land, the old model represents a situation where the scribe progresses through a given conceptual subdomain appearing in the omen compendium according to a route determined in advance. The scribe tries to walk across all the area in each subdomain, and it is clear when and how he should move from one subdomain to the next. Moreover, the path connecting all the subdomains is also determined in advance. The new model represents a situation where the scribe makes a heuristic walk through a given subdomain, taking opportunistic steps based on semantic salience and cross-domain overlaps. He does not try to cover all the area in a given domain but rather follows his nose within a limited range, ultimately switching to another subdomain under frame-shift.

Figure 3: Old versus new model of how the scribe constructs the Assyrian Dream Book.

Selected Bibliography

Coulson, S. 2001, Semantic Leaps – Frame-shifting and Conceptual Blending in Meaning Construction. Cambridge.
Fauconnier, G./M. Turner, 2002, The Way We Think: Conceptual Blending and The Mind’s Hidden Complexities. New York, NY.
Fillmore, C.J. 1982, ‘Frame Semantics’ in Linguistics in the Morning Calm. Seoul, 111–137.
Guinan, A. 1990, ‘The Human Behavioral Omens – On the Threshold of Psychological Inquiry’ in Bulletin of the Canadian Society of Mesopotamian Studies 19, 9–13.
Guinan, A. 2014, ‘Laws and Omens: Obverse and Inverse’ in J. Fincke (ed.) Divination in the Ancient Near East: A Workshop on Divination Conducted During the 54th Rencontre Assyriologique Internationale, Würzburg 2008. Winona Lake, IN, 105–121.
Oppenheim, L. 1956, The Interpretation of Dreams in the Ancient Near East -With a Translation of an Assyrian Dream Book. Transactions of the American Philosophical Society, Vol. 6, pt. 3. Philadelphia, 179–373.
Simonton, D.K. 2003, ‘Scientific Creativity as Constrained Stochastic Behavior – The Integration of Product, Person, and Process Perspectives’ in Psychological Bulletin 129.4, 475–494.

I wish to thank Elyze Zomer as well as Nicla De Zorzi and her team at the REPAC project for suggestions on the omen readings and interpretations given below. All errors remain my own.

4.4.‘If the Thunder Cries Like an Animal’

Horizontal Connections and Vertical Arrangement in EAE 44: 17-20

Written by Lucrezia Menicatti

How to cite: Menicatti, L., 2022, “If the Thunder Shouts Like an Animal, Version 01,” Project REPAC (ERC Grant no. 803060), 2019-2024, at (accessed day/month/year)

The discipline of astral divination involves the observation of heavenly bodies and celestial phenomena, whose conditions, appearance, and movements are interpreted as ominous signs. During the first millennium BCE, the omen texts devoted to this discipline were grouped and collected in a long series, Enūma Anu Enlil, “When Anu and Enlil” (EAE). In its standard version, the whole series consists of about 70 chapters, or Tablets, which were grouped into four major sections. These sections were already distinguished in antiquity according to the different celestial phenomena and heavenly bodies which they concerned: lunar phenomena, solar phenomena, meteorological and geological phenomena, planets and fixed star.

Tablet 44 of Enūma Anu Enlil belongs to the so-called Adad section of the astrological omen series, which considers storms and other weather phenomena and has been edited by Gehlken 2012: 11-34. Tablet 44 deals with the thunder and starts with a sequence of twenty omens whose protases describe the roll of thunder (dIŠKUR GÙ-šu) via simile as resembling the cry of various animals. The apodoses associate this sign with a prediction, whose content largely depends on the symbolic value that is attributed to the animal mentioned in the omen protasis.

The following analysis considers the four final omens of this sequence (i.e., EAE 44: 17-20). I address first the horizontal connections between protasis and apodosis in the individual omens, focusing especially on the symbolic values attributed to the animals in the protases. Such symbolic values influence outcome, topic, and setting of the corresponding predictions in the apodoses.

Second, I discuss the vertical connections between the omens. I examine the associations of animal names in the protases, and I show that graphic and phonological repetition enhances the perception of the sequence as a coherent unit. These connections between the animals’ names in the sequence of protases correspond to predictions that repeat and elaborate on the same theme in the apodoses. In other words, there is an attempt to create parallelism between the connections of the animals’ names in the protases’ sequence and the corresponding sequence of apodoses, as the following analysis shows.

The four omens read as follows:

šumma dAdad rigimšu kīma sisê iddi mātu šī ana panīša illak
“If Adad thunders like a horse – that land will make progress”.

šumma dAdad rigimšu kīma imēri iddi kurru āli iḫeppe maḫīru imaṭṭi
“If Adad thunders like a donkey – the kurru-dry measure of the city will be broken to pieces, business will decline”.

19) DIŠ dIŠKUR GÙ-šu GIM KUR.GImušen ŠUB SU.⸢GU7⸣ KUR i-maḫ-ḫar
šumma dAdad rigimšu kīma kurkê iddi sunqa mātu imaḫḫar
“If Adad thunders like a goose – the land will face famine”.

šumma dAdad rigimšu kīma sukannīni iddi mātu išdāša ana kaspi innaddin
“If Adad thunders like a turtledove – the land, its foundations will be given away for silver”.

Let us begin with the first omen of the sequence (17). The protasis compares the roll of the thunder to the horse’s neigh, and the corresponding apodosis forecasts that “the land will make progress”. The graphic repetition of the sign KUR in the protasis and in the apodosis establishes a link between the name of the horse, written logographically ANŠE.KUR.RA (akk. sisû), and the prediction concerning “the land”, also written logographically with the sign KUR (akk. mātu). The positive value attributed to the horse in Mesopotamia, which is a symbol of wealth and success (De Zorzi 2014: 158-159), influences the positive outcome of the prediction, which is the only positive forecast of the entire sequence.

This positive prediction is contrasted by the following one (18), which predicts that “the kurru-dry measure of the land will be broken to pieces, business will decline” (kurru āli iḫeppe maḫīru). The recession expressed in the two predictions (progress > decline) finds a parallel in the decrease of prestige between the two animals mentioned in the respective protases. In fact, although horse and donkey belong to the same family and relate to the same semantic domain, they symbolise quite different concepts. The horse is an extremely valuable animal and a luxury possession, while the donkey is mostly used for heavy works and thus represents inferiority and subordination (De Zorzi 2014: 159). The apodoses reflect this opposition. They concern the same topic (the land and its business) but predict opposite outcomes, namely progress and decline.

The repetition of the CVC sound group /kur/ in the Akkadian word kurru (spelled GUR-) also establishes a link between these two consecutive apodoses. This sound is reminiscent of the sign KUR, Akk. mātu, “the land”, which occurs in the preceding entry’s apodosis (17) as the subject. This suggests a link between this prediction and the same semantic domain.

The following two entries’ protases (19-20) build again on the repetition of the same sign KUR and on the phonological repetition of /kur/. The sign KUR occurs in the name of the goose (KUR.GI7), while its Akkadian spelling, kurkû, repeats the sound group /kur/. Also, the name of the turtledove (TU.KUR4, sukannīnu) in the following entry includes the sign KUR4, homophonous of KUR.

Birds are generally associated with famine and destruction in Mesopotamian divination (De Zorzi 2014, 166), and the mention of these two types of birds corresponds to two negative predictions in the apodoses. Moreover, the apodoses elaborate once more on the same graphic repetition involving the sign KUR, which occurs in both the apodoses with the reading mātu, “land”. The apodosis in 19 predicts that “the country will experience famine” (sunqa mātu imaḫḫar). The alliteration of sibilant /s/ and nasal /n/, velar /q/ and /k/, and the consonance of /u/ and /a/ between the term SuNQa, “famine” (19), and SuKaNNīNu “turtledove” (20) establishes a phonological link between this apodosis and the following entry’s protasis. The apodosis of this entry (20) predicts that “the foundation of the land will be given away for silver” (mātu išdāša ana kaspi innaddin), thus elaborating on the same theme of famine and impoverishment.

Therefore, the apodoses of these four omens in EAE 44: 17-20 build consistently on the same theme: they all concern the land and its wealth. The repetition of the same sign KUR and of the same CVC group of sounds /kur/, both in the protases and in the apodoses, reinforces this thematic coherence. This graphic and phonological repetition relates the animal names occurring in the protases, starting from the largest and most prestigious horse (ANŠE.KUR.RA), which relates paradigmatically to the smaller and much less prestigious donkey (ANŠE), to the goose (KUR.GI7), and finally, to the smallest, the turtledove (TU.KUR4). These animal names in the protases’ sequence trigger predictions concerning the land, mātu, which is also written logographically by the sign KUR.

Also, the regressive order in the dimensions of the animals in the protases corresponds to a sequential arrangement in the apodoses’ sequence. The sequence starts by predicting progress for the land (17). Then, due to inflation (the kurru-dry measure is said to “be broken to pieces”), business declines (18). The decline of the land’s business leads to a famine (19), which ultimately results in the necessity of selling away the land itself for money (20). Thus:

Other textual elements tie the beginning of the sequence to its end. The contrast between panu, “front”, in the first apodosis of the sequence (17), and išdu, “base”, in the final entry’s apodosis (20) reinforces the opposition between the first prediction forecasting “progress” and the final one predicting that “the land will be given away for silver”. Similarly, the active action that the land performs in the first entry’s apodosis (lit. “the land will go towards its front”) contrasts with the passive action which the land undergoes in the final entry’s apodosis (“the land will be given away for silver”). These oppositions reinforce the perception of the apodoses’ sequence as a thematic unit, framed by the only positive and the most negative predictions.

In conclusion, we can see that the four omens at the end of the “animal sequence” in EAE 44: 17-20 reveal an attempt to create a block of four inter-connected omens. Our textual analysis shows the role of repetition as a structural device in the arrangement of this sequence. Phonological, graphic, and semantic repetitions create a consistent sequence of omens in which the connections between the animal names in the protases correspond to thematically homogenous predictions in the apodoses.

Selected Bibliography
De Zorzi, N. 2014, La Serie Teratomatica Šumma Izbu: Testo, Tradizione, Orizzonti Culturali (2 Volumes). History of the Ancient Near East / Monographs – XV, Padova.

Gehlken, E. 2012, Weather Omens of Enūma Anu Enlil: Thuderstorms, Wind and Rain (Tablets 44-49). Cuneiform Monographs 43, Leiden/Boston.

4.5. ‘From White, to Red, to Dark’
Colors and Their Interpretation in Manzāzu 2: 42-45

Written by Lucrezia Menicatti

How to cite: Menicatti, L., 2023, “From White, to Red, to Dark, Version 01,” Project REPAC (ERC Grant no. 803060), 2019-2024, at (accessed day/month/year).

The vast majority of Babylonian divinatory texts includes long lists of omens structured as conditional clauses. The omen protasis gives a certain sign, and the apodosis provides the corresponding interpretation. The connection between sign and prediction within single omens – the ‘horizontal connection’ in REPAC’s terminology – is based on some similarity between the two on the graphic, phonological, or semantic level (De Zorzi 2022: 376-377). Graphic or phonological similarities may overlap with semantic relationships, thus creating multi-layered connections within a single omen. In sequences of omens, the same principles connect the protases and the apodoses of different omens on the vertical level, i.e., they create ‘vertical connections.’

This showcase examines such horizontal and vertical connections in a sequence of four omens from the divinatory series Bārûtu, “The diviner’s craft,” which is devoted to the discipline of extispicy, namely, the divinatory inspection of the liver and other organs of the sacrificial sheep. Cuneiform tablets dating to the VIII-VII centuries BCE coming mostly from the libraries of Ashurbanipal in Nineveh (modern Kuyunjik in northern Iraq) attest the so-called ‘standard version’ of this divinatory series, which consists of ten major chapters, each concerning the outer appearance of the sacrificial sheep, various parts of the entrails, or specific ominous features observed on the exta. These ten chapters are further divided into about one hundred sub-chapters, or Tablets.

Our omen sequence belongs to the third chapter of Bārûtu. The chapter is dedicated to the ‘Presence’ (manzāzu), which is the name given to a vertical groove on the left lobe of the sheep’s liver. This section of the liver is often divided into three sub-sections, namely, Top, Middle, and Base (Koch 2000: 52-53). Our omen sequence belongs to the second sub-chapter, or Tablet, of Manzāzu (Manzāzu 2: 42-45) and concerns the occurrence of a special ominous sign, called the ‘Request-mark’ (erištu), in the Presence. This mark, for which different identifications have been suggested (Koch 2000: 48), is probably to be identified with a small “formation of fat, shaped like a seed” (Jeyes 1989: 86) that may occur on the liver.

The first omen in our sequence (42.) observes the finding of two Request-marks in the Presence, likely in its Top. More specifically, the protasis refers to a second Request-mark lying on top of a Request-mark. In the following omens (43.-45.), different colors describe the appearance of the Request-mark that lies on top: white, red, and, finally, dark. The color shading tarku, which we translate as ‘dark,’ probably identifies shades of dark blue or purple (Thavapalan 2020: 163) and frequently concludes similar sets of colors in omen sequences from Old Babylonian extispicy texts (Winitzer 2017: 357-359).

In this showcase, we demonstrate that the apodoses of these four Manzāzu omens build a sequence of predictions that is meant to reflect the paradigmatic set of colors in the protases in a parallel fashion. To do this, we first address the horizontal connections between protasis and apodosis within the individual omens of the sequence under investigation. Second, we analyze the vertical arrangement of the sequence in order to determine the role played by the color variables in structuring the sequence of protases and their relationship with the sequence of predictions in the apodoses.

Two tablets from Nineveh, both kept at the British Museum, preserve our omen sequence: K 2258 + K 3271, and K 9991. K 2258 + K 3271 is missing a fragment at the bottom left corner that presumably included the beginning of these lines (lines 23’-26’ on the tablet). K 9991 is a small fragment that breaks after line 10’, corresponding to omen 43’ of this sequence. We quote the omen sequence following Koch’s (2000: 88-89) edition of these two tablets. A few more signs have been integrated following the most recent reading of K 2258 + K 3271 which is available at

1. No Color

The first omen’s protasis (42.) describes the observation of a second Request-mark lying above a Request-mark. The beginning of the protasis is not preserved. However, we follow Koch in assuming that these marks are found in the Top of the Presence (rēš manzāzi), since all of the omens preserved on this Tablet of Manzāzu (see above) deal with that sub-section.

In Mesopotamian extispicy, the Presence symbolizes the god’s presence in the divinatory liver; therefore, its absence represents a very negative sign, meaning that the gods do not grant their support and do not wish to accept the sacrifice (Jeyes 1989: 53-54; Winitzer 2010: 186-192; Menicatti 2022: 228-230)

In extispicy omens, the Request-mark (erištu) is frequently associated through paranomasia with apodoses predicting a request (erištu) of some kind (Jeyes 1989: 86; Winitzer 2017: 108-109). The agent placing the request is usually a deity. See, for instance, this Old Babylonian extispicy omen: šumma ina rēš naplaštim erištum erišti ilim rabîm ša ginêm ilum irriš, “If at the Top of the View (there is) a Request mark – a request of the great god: the god requests a regular offering” (AO 9066: 26-28; Winitzer 2017: 110).

The Request-mark thus usually represents a request from a deity, and its correct interpretation implies the correct interpretation of the god’s wishes. This mark can thus be considered a positive sign; when it occurs on the liver, it represents the manifestation of the god’s demand. At the same time, the abnormal appearance of the Request-mark is a negative sign: it indicates that there has been some disturbance in the interaction with the divine world. The associated apodoses predict that the god’s demand cannot be interpreted or, for some reason, cannot be met. See, for instance, this Old Babylonian extispicy omen (AO 7029 obv. 3-5; Jeyes 1989: 86): šumma ina muḫḫi ṣibtim erištim tarkat erišti ina bīti awīlim innerrešu ul innaddin, “If a Request-mark is dark at the crest of the Increase – the demand that will be raised in the man’s house will not be met.” Here, a dark coloration of the Request-mark symbolizes the failure to fulfill the god’s wish.

In our omen, there are two Request-marks, one lying on top of the other. The doubling of a zone or a mark generally represents a negative sign in Mesopotamian divination, symbolizing abnormality and ambiguity (De Zorzi 2014: 186-187; Winitzer 2017: 411-412). The two Request-marks here may therefore represent an ambiguous request, which cannot be understood and thus cannot be fulfilled accordingly. The image of the second Request-mark lying on top of a Request-mark and, implicitly, covering it seems to reinforce this idea: a second Request-mark conceals the first Request-mark, making it impossible to distinguish and interpret the divine demand.

This image results in a negative apodosis predicting “divine punishment” (šērtu). Phonological repetition strengthens the semantic link between the Request-mark and its symbolism in the protasis and the “divine punishment” in the apodosis. The two Akkadian terms (eRiŠTu <—-> ŠēRTu) repeat the same vowel sounds /e/ and /u/ and alliterate /š/, /r/, and /t/. Erištu and šērtu in fact contain the same consonants, only with /r/ and /š/ in reversed order (ršt <—-> šrt). This can be regarded as an instance of what Noegel (2021: 273) calls “anagrammatic paranomasia” – namely, two words containing the same consonants, but in a different order. This device is attested in many other contexts in Akkadian written production, as Noegel (2021: 273-275) points out.

Thus, while the normal appearance of a Request-mark in a section of the liver generally corresponds to a divine request (eRiŠTu <—-> eRiŠTu), the abnormal appearance of this mark results in a prediction involving divine punishment (šērtu). The inversion seems to work on the semantic as well as on the phonological level (eRiŠTu <—->ŠēRTu).

Visual similarity may establish a further connection between these two items. The word šērtu indicates a sickness, understood as the god’s punishment for some offense or sin. Šērtu occurs also in other contexts and seems to refer to dropsy or to some skin diseases like leprosy or boils (Howe 2021: 67-68). As the Request-mark is probably a small protrusion of fat on the liver, I would argue that this mark may visually evoke skin diseases, or the swellings caused by dropsy.

The apodosis of omen 42. predicts that a sinner, who has transgressed the will of the god and bears the visible stigmas of the divine anger upon him (šērtu), will repeatedly carry his impiety into the palace. The choice of the Gtn stem verb in the apodosis (ītenerreb, “he will keep entering”) may also reflect the image of the two Request-marks lying one on top of the other described in the protasis, which conveys reiteration. Also, the setting of the apodosis in the palace’s sphere probably depends on the occurrence of the negative sign in the sub-section of the Top of the Presence, which can represent metaphorically the top of political power in extispicy texts (Menicatti 2022: 226-230; 238-239).

The following table summarizes the horizontal connections between protasis and apodosis in this omen:

2. No Color > White

The next entry’s (43.) protasis reads, [— eli erišti erišt]u nadât u peṣât, “[—- a Request] lies [on top of a Request] and it is white.” The addition of the variable white in this protasis with respect to that of the preceding omen corresponds to a parallel addition in the apodosis, which implicitly refers to the preceding one (ditto), but also adds a new prediction, namely, that the sick person entering the palace will be caught by his ibru. Koch translates ibiršu as “his friend.” CAD I/J: 6-7 s.v. ibru, however, suggests the translation of ibru as “friend” only in contexts that clearly refer to two people having some emotional relationship with one another like, for instance, Gilgamesh and Enkidu in the Epic of Gilgamesh. For this reason, following CAD I/J: 7, s.v. ibru, we instead translate ibiršu here as “his peer.” We understand this term as denoting someone belonging to the same social status and environment of the “one who is afflicted by divine punishment.”

The following table highlights the correspondences between protasis and apodosis in this omen:

I would argue that it is the addition of the color variable white in the protasis of 43. that determines this shift from a fully negative apodosis – a sick person keeps entering the palace (42. and 43a.) – to a non-negative one – this person will be caught (43b.). It has been demonstrated that white in Mesopotamia is often associated with the idea of emptiness and deprivation (Thavapalan 2020: 140). In our divinatory context, white seems to represent the annulment or subversion of the preceding omen’s negative charge.

3. White > Red > Dark

In the following omen (44.), the Request-mark is described as red ([—- eli erišti erištu] nadât u sāmat “[—- a Request] lies [on top of a Request] and it is red.” The apodosis refers implicitly to 42a. (ditto) and then repeats 43b. verbatim, i.e., it predicts that the sick person who illegitimately enters the palace will be caught by his peer (ibiršu iṣabbassu). Finally, it adds a new prediction, namely, the sinner “will be burnt” (iqqallu). The color red in the protasis evokes the image of fire, thus triggering the horizontal connection with this prediction.

The last omen’s protasis (45.) is almost entirely broken. K 2258 + K 3271 only preserves the sign -at according to the reading on This sign probably represents the ending of the feminine stative form tarkat, “it is dark.” As mentioned in the introduction, this color often occurs at the end of similar sequences of colors in Babylonian extispicy compendia (Winitzer 2017: 357-359). Furthermore, the color term tarku is consistently associated with death in extispicy and other contexts (Jeyes 1980: 112; Winitzer 2017: 241-243; Thavapalan 2019: 162-164). Since this omen’s apodosis predicts death, I would suggest that the protasis attributed this color variable (tarkat) to the Request-mark. The apodosis in fact repeats that of omen 44. but predicts that the sick person “will be killed” (iddâk, 45.).

The following table highlights the correspondences between the colors mentioned in the protases and the sequence of predictions in the apodoses:

If our reconstruction of the final omen’s (45.) protasis is correct, then the protases would include a color sequence starting from no color > white > red > dark. The apodoses reflect this scheme, but not symmetrically. The addition of the variable “white” in 43. corresponds to an expansion of the prediction introduced in 42., an expansion which in fact reverses the polarity of the prediction.

The following two omens replace “white” with “red” and “dark,” respectively. Their apodoses add new predictions that reflect the different colors mentioned in the protases, but they both also repeat the same prediction from 43. – “ditto, but his peer will seize him” – even though the corresponding protases lack the color variable white

In other words, the paradigmatic scheme of colors in the protases’ sequence corresponds to a parallel system of variation in the apodoses’ predictions, which create a short narrative reflecting the color gradation in the protases. In the apodoses of 44. and 45., the logic of this narrative overcomes the correspondence with the color variables in the protases. The two apodoses incorporate the prediction from 43. – the sick person entering the palace will be caught – which is the necessary premise for the two forecasts introduced in 44. and 45., namely, that the sick person will be burnt, and, finally, killed.

The predictions thus build on a narrative climax. The first prediction is negative and forecasts that a sick man will keep entering the palace (42.). In 43., this person will be caught by his peer. 44. not only predicts that this man will be caught by his peer, but it also goes one step further and predicts that he “will be burnt.” Finally, the apodosis of 45. concludes the apodoses’ sequence with the explicit mention of this man’s death.

These findings allow two considerations. On the one hand, the analysis of Manzāzu 2: 42-45 shows the attempt at structuring the sequence of protases on the basis of a color paradigm Ø>white>red>dark. The apodoses modify the predictions in correspondence with the color variables, thus creating a parallel sequence of predictions reflecting the color scheme in the protases.

On the other hand, 44. and 45. put the coherence of the predictions’ sequence before their symmetric arrangement in correspondence with the protases. This shows that there is a scheme of correspondence between protases and apodoses in these omens; the scheme, however, is not fixed and allows a certain freedom.

Selected Bibliography

De Zorzi, N. 2014, La Serie Teratomatica Šumma Izbu: Testo, Tradizione, Orizzonti Culturali (2 Volumes). History of the Ancient Near East / Monographs – XV, Padova.

De Zorzi, N. 2022, ‘Parallelism and Analogical Thought in Babylonian Poetry – Case Studies from Ludlul bēl nēmeqi, the Babylonian Theodicy, and the Šamaš Hymn’ in Wiener Zeitschrift für die Kunde des Morgenlandes 112, 367-394.

Howe, A. 2021, Conceptions of Transgression and Its Consequences in the Mesopotamian Exorcistic Corpus. Unpublished PhD Dissertation, Wolfson College.

Jeyes, U. 1980, ‘Death and Divination in the Old Babylonian Period,’ in A. Bendt (ed.) Death in Mesopotamia – Papers read at the XXVI Rencontre assyriologique international. Copenhagen, 107–121.

Jeyes, U. 1989, Old Babylonian Extispicy. Omen texts from the British Museum. Publications de l’Institut historique-archéologique néerlandais de Stamboul 64. Istanbul.

Koch, U. 2000, Babylonian Liver Omens – The Chapters Manzāzu, Padānu and Pān Tākalti of the Babylonian Extispicy Series mainly from Aššurbanipal’s Library. The Carsten Niebuhr Institute Publications 25. Copenhagen.

Menicatti, L. 2022, ‘Top – Middle – Base. A System of Omen Sequencing and its Interpretation in Barûtu Chapter 3 and in Old Babylonian Precursors’ in Wiener Zeitschrift für die Kunde des Morgenlandes 112, 223-255.”

Noegel, S. 2021, “Wordplay” in Ancient Near Eastern Texts. Ancient Near Eastern Monographs 26, Atlanta, GA.

Thavapalan, S. 2020, The meaning of color in ancient Mesopotamia. Culture and History of the Ancient Near East 104, Leiden / Boston, MA.

Winitzer, A. 2010, ‘The Divine Presence and Its Interpretation in Early Mesopotamian Divination’ in A. Annus (ed.) Divination and the Interpretation of Signs in the Ancient World. Chicago, IL, 177-197.

Winitzer, A. 2017, Early Mesopotamian Divination Literature – Its Organizational Framework and Generative and Paradigmatic Characteristic. Ancient Magic and Divination 12, Leiden / Boston, MA.

4.6. Repetition as an Argumentative Technique in the Diagnostic Handbook

Written by Eric Schmidtchen

How to cite: Schmidtchen, E., 2023, “Repetition as an Argumentative Technique in the Diagnostic Handbook, Version 01,” Project REPAC (ERC Grant no. 803060), 2019-2024, at (accessed day/month/year).

I Introduction: The Diagnostic Handbook

Recent research on Mesopotamian divinatory texts has demonstrated that repetition could take on the role of a intentionally applied technique to create connections between elements of omen entries’ protases (antecedent) and apodoses (consequent) and for sequences of entries (although the underlying associative or symbolic ties are not always obvious to modern scholars). In addition, repetition was also used to create a progression of these connections in certain omen entries and sections (see the REPAC showcases 4.1–4.2 = de Zorzi 2020; Menicatti 2021 and 4.4–4.5 = Menicatti 2022; Menicatti 2023). In the present showcase, it will be demonstrated that different forms of repetition in Mesopotamian divinatory texts could serve as an argumentative technique in order to make an assertion or proposition not just at the level of limited clusters of entries but also at the level of the text’s macro-structure, that is for an entire tablet or series and even beyond, viz. to serve as a ‘generalising’ statement.

The relevant examples stem from the prognostic-diagnostic standard series Sakikkû (also referred to as the Diagnostic Handbook; see Heeßel 2000 = BAD in general and pp. 150–170, in particular, for an edition of tablet 15; cf. also Schmidtchen 2021 = BAM 13: 25 and 623 for the tablet’s beginning). The series is attested mainly from the 1st millennium B.C. However, its standardisation by the scholar Esagil-kīn-apli from Borsippa (see Finkel 1988: 144f.; Schmidtchen 2018a) can be traced back to the end of the 2nd millennium. It comprises 40 tablets which are structured into 6 subseries, or chapters, whose titles are in turn always mentioned separately in the tablet’s colophons. Besides the structure, the Tablet numeration of these tablets remained fairly stable until the second half of the 1st millennium which is why this redaction is often called ‘canonical.’

Each of the series chapters has a distinct focus. The 1st chapter (Tablets 1–2) presents the incantation expert (āšipu) on the way to a sick man’s house. On the way, he scrutinises various significant phenomena which may have relevance for the patient’s condition. In chapter 2 (Tablets 3–14), the patient is examined “from head to foot” (ištu muḫḫi adi šēpī, the a capite ad calcem-structure, is explicitly mentioned in Esagil-kīn-apli’s editorial note; see Finkel 1988:148f. A 61–62 // B 25ʹ; Schmidtchen 2018b: 317 line 61)—an ordering principle which is frequently found in other divinatory texts for observations on the human body. Chapter 3 (Tablets 15–25) considers the long-term development of, stages of, and side effects of diseases and sicknesses. This includes the span of days the patient has been sick, the particular symptoms of fever and evacuation, digestion, diet and so on. The series’s overall perspective changes in chapter 4 (Tablet 26–30) from examining phenomena and the patient with the intent to gain information on the nature of a disease to concentrating on distinct sets of maladies associated with demons and ghosts such as stroke, epilepsy, or sicknesses with similar symptoms. This focus on an already-known disease or ailment continues in chapter 5 (Tablets 31–35) which also contains the most connections to Mesopotamian medico-therapeutic texts. Chapter 5’s first Tablets deal with different forms of fever or ailments associated with transmission by the wind. Additionally, the chapter includes diagnostic lists for different kinds of diseases (mainly skin diseases and diseases associated with a person’s locomotor system) combined with the so called šumšu-formula—a descriptive pattern in the form of “If the appearance of (stone/plant etc.) is (like) …: X is its name”, used to describe such things as stones, plants, snakes and also diseases. The chapter also covers the treatments of lesser and more severe forms of human-induced magic including witchcraft. The 6th and final chapter (Tablets 36–40) focuses on observations of the female body, especially in connection with pregnancy, its course as well as on the ailments of infants.

In contrast to other divinatory series, the Diagnostic Handbook occupies a special position within both the divinatory and medical corpus of Mesopotamian erudite texts since it combines divinatory systematology (chapters 1–3, 6)—such as the use of associative, phonemic or graphemic connections between an omen entry’s protasis and apodosis or the systematic use of polarities (such as right and left or up and down but also polar semantics of several verbs)—with magico- and medico-therapeutic content (mainly chapters 4–5), like prescriptions for short rituals and treatments or the preparation of materia medica/magica.

The main purpose of the series is to present information on the patient’s condition (prognosis), on an illness (diagnosis of diseases and illnesses), on an illnesses’ originators (or causees; again, a diagnosis, viz. direct disease-causing agents like gods, ghosts, demons, and witchcraft etc.) and in certain cases, also its causes (e.g. by aetiological indications). These were, for example, the loss of protection provided by the patient’s protective deities either because the patient broke an oath or because he became otherwise polluted through bad or external influences such as witchcraft. The Diagnostic Handbook thus fulfilled a ‘medical’ function, in general, by providing the responsible expert with information about the possibility of a patient’s recovery and with information about the particularities of a patient’s treatment with the help of systematic considerations and techniques otherwise known from divinatory texts sensu stricto.

For this showcase, we will look more closely at the 15th Tablet of the Diagnostic Handbook. This Tablet is also the beginning of the handbook’s 3rd chapter which immediately follows the 2nd chapter’s a capite ad calcem-observation of the patient (cf. also II.3. below for this ordering principle’ relevance for the showcase).

II Repetition(s) in Sakikkû Tablet 15

How is the argumentative function in said text expressed with regard to repetition in order to pose a ‘generalising’ statement or proposition? To begin with, it is crucial to note that the relevant repetitions within Sakikkû Tablet 15 represent different types thereof:
(1.) The first type of repetition concerns the prognoses within the entries’ apodoses (47 out of 99 preserved entries), which all have the same negative prognosis (mostly written with the logogram GAM for imât “he will die”; cf. the GREEN area in fig. 1 below). Extrapolating from parallel omens found in other sections of the Diagnostic Handbook, a negative outcome can likewise be assumed with some certainty for the fragmentarily preserved entries (cf. also II.3. below).

These negative prognoses repeat the exact same verbal form (“he will die”; viz. exact repetition) which indicate that all of the significant signs (or symptoms) mentioned within each entry’s protases are to be understood as indications (see II.2. below) of the serious development of the affliction and the death of the patient.
(2.) The second type of repetition is more subtle but nevertheless recognisable. The main bulk of symptoms described in the protases of Tablet 15 can, with a few exceptions, be subsumed into roughly four core groups (see BAM 13: 221):

In connection with the first type of repetition (the exact repetition of the verbal form imât), these repeated bodily signs are thus to be seen as particularly significant indications for the above-mentioned negative apodoses. However, different from (II.1.), these repeated signs are not always exact repetitions (although some symptoms are often repeated verbatim) but are repetitions of (respective to each group) semantically-connected states and events. At this point of our assessment, these repetitions constitute a negative evaluation or interpretation of each mentioned symptom group (II.2.a–d) valid for the entirety of Tablet 15.

(3.) Two additional types of repetition in Sakikkû Tablet 15 are connected in particular with the Tablet’s and the series’s structure.

a) This Tablet includes a large number of omen entries repeated verbatim from other parts of the Diagnostic Handbook, in particular from the a capite ad calcem-structured 2nd chapter, as for example shown in the following two entries:

Ex. 8: DIŠ ina GABA-šú SIG-iṣ-ma MÚD ú-tab-ba-ka u ú-rap-pad ŠU dU.GUR GABA.RI SIG-iṣ GAM
If he is affected in his chest so that he expectorates (lit. pours out) blood and is restless – hand of Nergal; he is affected on the front; he will die.

(Sakikkû 12: 3)

Ex. 9: DIŠ KI.MIN(UD 1.KÁM GIG)-ma ina GABA-šú SÌG-ma MÚD ú-tab-ba-ka u ú-rap-pad ŠU dU.GUR GABA.RI ˹SÌG˺-iṣ GAM
If ditto (he is sick for one day) and he is affected in his chest so that he expectorates (lit. pours out) blood and is restless – hand of Nergal; he is affected on the front; he will die.

(Sakikkû 15: 38′ / BAD 15: 32′)

In addition, the introduction of tablet 15 “If he is sick for/since once day and …” is repeated from the second entry onward with the signs KI.MIN “ditto” which essentially expands each subsequent omen and emphasises the severity of the symptoms’ temporal continuation.

b) Tablet 15 not only repeats certain entries from other sections of the series, but it also resumes the structural principle a capite ad calcem of Sakikkû chapter 2 in its first section (1–73′). The subsequent section (74′–99′) is, as far as preserved, concerned with similar symptoms regarding the entire body and general anomalies. This latter section might refer back to the structure of Sakikkû chapter 3, which also enlarges the scope of the patient’s examination to include additional categories like time, fever, intake and expectorating, nutrition and so forth (see I).

I propose for the already-mentioned points that the repetition types a) (repetition of omen entries/parallels) and b) (macro-structural repetition) thus expand the argument for (II.1.) the overall severity and deadly outcome of (II.2.) the repeatedly-mentioned four symptom groups beyond Sakikkû Tablet 15. This is achieved by first referring structurally (viz. to the a capite ad calcem-principle) to the main part of the patient evaluation which occurs in the divinatory-influenced chapter 2 (and to some extant chapter 3 as well). Secondly, this frame of reference is sustained by continuously and explicitly repeating (or paralleling) parts of chapter 2.

III Repetition as an Argumentative Technique for Generalisation?

Another indirect indication underlines the generalising notion of the repetitions presented above. Nearly all of the entries in the Diagnostic Handbook (including those from chapters 4–6) in which at least one or more of the four symptom groups (see II.2. above) are presented have a likewise negative prognosis (see BAM 13: 219ff.) even though they are not mentioned within Tablet 15. That these symptom groups are repeatedly referenced is thus most likely meant to emphasise the lethality of these signs (at least within the scope of the Diagnostic Handbook) and to easily transfer them quasi ‘inductively’ onto other sections of the series and beyond.

A quite explicit reference is given in the Sakikkû-catalogue by its editor Esagil-kīn-apli (see Finkel 1988: 146 A 20; BAD: 14ff.; BAM 9: 137–157. 313–333), as well as in the serial tablet-rubrics of chapter 3 whose title is represented by the incipit of Tablet 15 in combination with an editorial remark (which originally was part of Tablet 15’s incipit-line). The 3rd chapter’s title is as follows:

DIŠ UD 1.KÁM GIG-ma šá TAG-ti
(Tablet … of) ‘If he is sick for one day and’ (omens/symptoms) which are ill-portending

(Sakikkû catalogue Ms. A 20′; rubrics of Tablets 16, 17, 22; var. tablet 23)

Although the phrase šá/NÍG TAG-ti has been formerly interpreted as “das Aussehen der Berührung” (BAD: 16), the variant formulation šá TAG-tú-ti in the rubric of tablet 23 indicates the signs should be read as ša laptūti “that which is ill-portending/anomalous” (BAM 9, 140f.; cf. for this meaning of laptu CAD L: 95f. s.v. laptu 2 and CAD Š/1: 259 s.v. šalmu 1e).

This short remark refers thus to the fact that the signs or symptoms presented within Sakikkû Tablet 15 are indeed leading unanimously to fatal outcomes and that they are intentionally presented as such. Furthermore, together with the different forms of repetition (see II.1.–3. above), they constitute an argument that expands the validity of the interpretation of said symptom groups over the series’s entirety.


Heeßel, N. P. 2000, Babylonisch-assyrische Diagnostik. Alter Orient und Altes Testament 43, Münster (= BAD).

Schmidtchen, E. 2018a, ‘Esagil-kīn-apli’s Catalogue of Sakikku and Alamdimmu’ in U. Steinert (ed.) Assyrian and Babylonian Scholarly Text Catalogues. Medicine, Magic and Divination. Die babylonisch-assyrische Medizin in Texten und Untersuchungen 9, Berlin/Boston, 137–157 (= BAM 9).

Schmidtchen, E. 2018b, ‘The Edition of Esagil-kīn-apli’s Catalogue of the Series Sakikkû (SA.GIG) and Alamdimmu’ in U. Steinert (ed.) Assyrian and Babylonian Scholarly Text Catalogues. Medicine, Magic and Divination. Die babylonisch-assyrische Medizin in Texten und Untersuchungen 9, Berlin / Boston, 313–333 (= BAM 9).

Schmidtchen, E. 2021, Mesopotamische Diagnostik: Untersuchungen zu Rekonstruktion, Terminologie und Systematik des babylonisch-assyrischen Diagnosehandbuches und eine Neubearbeitung der Tafeln 3–14. Die babylonisch-assyrische Medizin in Texten und Untersuchungen 13, Berlin/Boston (= BAM 13).

Finkel, I. L. 1988, ‘Adad-apla-iddina, Esagil-kīn-apli, and the Series SA.GIG’ in E. Leichty, M. de Jong Ellis and P. Gerardi (ed.) A Scientific Humanist. Studies in Memory of Abraham Sachs (FS Sachs). Philadelphia, 143–159.


Menicatti L. 2021, ‘Stealing the Property of the Gods – Observations on a Top-Middle-Base Omen Sequence from an Old Babylonian Liver Model’, Version 01, Project REPAC (ERC Grant no. 803060), 2019-2024, at DOI: 10.25365/phaidra.303 (accessed 04/08/2023)

Menicatti L. 2022, ‘If the Thunder Shouts Like an Animal’, Version 01, Project REPAC (ERC Grant no. 803060), 2019-2024, at (accessed 04/08/2023)

Menicatti L. 2023, ‘From White, to Red, to Dark’, Version 01, Project REPAC (ERC Grant no. 803060), 2019-2024, at (accessed 04/08/2023).

De Zorzi, N. 2020, ‘“If a pig carries a reed and enters a man’s house …” Observations on Some Structuring Devices in Babylonian Omen Lists’, Project REPAC (ERC Grant no. 803060), 2019-2024 at DOI: 10.25365/phaidra.230 (accessed 04/08/2023).