Citatio: B. Christiansen (ed.), hethiter.net/: HDivT (07-05-2026)

Bird Oracles (MUŠEN)

1. Short Description

The observation of free-flying birds in a concrete observation field is one of the most important and best documented divinatory techniques the Hittites employed to interrogate the gods on matters of great importance, such as success in war, health, cultic issues, and royal succession. To modern readers, these texts initially seem confusing due to their complex technical language and repetitive nature. However, upon closer examination, they reveal themselves to be remarkably interesting documents that are unique in the ancient world. Not only do they provide a wealth of information about historical events, social structures, and religious practices, but they also demonstrate a highly structured reasoning and a formalized empirical procedure. This is evident in the detailed account of the observations, the careful and precise questions, the use of specialized vocabulary, the stenographic style—characterized by numerous abbreviations and cursive writing—and the communication of divine verdicts through the augurs, who are frequently mentioned by name. Furthermore, the outcome of one procedure is often cross-checked by another bird oracle or a different technique. Particularly remarkable is the detailed and highly formalized report of the bird observation. The texts document which birds appeared in the observation field, from which direction they came, where they flew, how they perched, called, fought, left droppings, turned their beaks, or disappeared. As an example for such an observation, a passage of a bird inquiry concerning the king’s and queen’s winter stay and the celebration of various festivals at Ḫattua, is cited in the following (KUB 18.12 obv. 7–14, CTH 564):

An eagle was perching tar(uya)li-. Its beak, however, was turned diagonally (pa(riyaw)an). A ḫarrani- bird, however, was perching GU[N]-li-. Its beak, however, was turned away from the middle. An aliliya- bird, however, came in front from the favorable side and it flew away in the middle. An aršintatḫi- bird, however, [was] per[ching] tar(uya)li-. Behind the road: An aramnant- bird came tar(uya)li- diagonally. On the second day: Four ḫarlitinzi- birds: tar(uya)li-. An eagle, however, [was perching] GUN-li-. An(other) eagle, however, came in front from the favorable side and it flew away in the middle. A ‘fox-bird,’ however, flew tar(uya)li- diagonally. Behind [the road]: [...] came diagonally. On the third day: An aliliya- bird flew tar(uya)li- diagonally. An eagle, however, was perching GUN-li-. Its beak, however, was [...]. We conjured. An aliliya- bird came in front from the favorable side. And it flew away in the middle. A ‘wide-wing’ bird […]. Behind the road: A ḫarrani- bird came tar(uya)li- diagonally. Thus (speak) Piḫatarḫunta and Armana[ni]: “It was confirmed.”

2. Research History

2.1 Discovery and First Cuneiform Copies

The textual basis for the study of Hittite bird oracles was laid by the excavations at Boğazköy (ancient Ḫattuša) conducted by H. Winckler and Th. Makridi in 1906–07, which yielded more than ten thousand clay tablets, including many oracle texts. Handcopies were published in the series Keilschrifttexte aus Boghazköi (KBo), whose first volumes appeared in 1916 (KBo 2) and 1920 (KBo 1). KBo 2 already contains two handcopies by H. Figulla of two large oracle texts concerning Tutḫaliya’s IV accession to the throne (KBo 2.2) and the Armatarḫunta and Šaušgatti affair (KBo 2.6), the latter of which could later be joined with two additional fragments (KUB 18.51 and KUB 50.5).Between 1922 and 1928, A. Walther published five volumes dedicated to oracle texts in the second major series of Hittite cuneiform texts, Keilschrifturkunden aus Boghazköi (KUB), namely KUB 5, KUB 6, KUB 16, KUB 18, and KUB 22. A substantial collection of bird and other oracle texts was published by A. Archi between 1979 and 1983 in KUB 49, KUB 50, and KUB 52. Further bird oracle texts found in Boğazköy, Mașat Höyük (Tapikka), Kayalıpınar (Šamuḫa), Kușaklı (Šarišša) appeared in later KBo and KUB volumes as well as other series.

2.2 Early Reception: Dismissal and Neglect

Early Hittitologists, such as J. Friedrich and F. Sommer, held a very negative view of Hittite oracle texts in general, and bird oracles in particular. They even doubted the sense of publishing handcopies of them and openly showed their disapproval of Figulla’s and Walther’s meticulous efforts to publish these texts (Friedrich J. 1926f; Friedrich J. 1927f; Sommer F. 1924b; Sommer F. 1930c). These judgments contributed to the scholarly neglect of oracle texts – and bird oracles in particular – for decades.

2.3 First General Treatments and Early Editions

The first short overview of Hittite oracle types was provided by A. Goetze in his groundbreaking monograph “Kleinasien,” published in 1933. He identified the three main techniques: extispicy (SU), bird oracles (MUŠEN), and symbol oracles (KIN), and highlighted the significance of oracle inquiries in Hittite culture (Goetze A. 1933a, 148–150). A. Boissier followed with a comparative study of Hittite and Babylonian oracles, taking also Etruscian and Roman practices into account (Boissier A. 1935a).

2.4 Laroche’s Catalog and Classification

The significant organizational achievement of the corpus was E. Laroche’s “Catalogue des textes hittites” (CTH), which was published in several stages between 1956 and 1975 (Laroche E. 1956h; Laroche E. 1957c; Laroche E. 1958e; Laroche E. 1971c; with supplements in RHA 30, 1972, 94–133, and RHA 33, 1975, 68–71). For bird oracles, he designated the following numbers: CTH 573 (pure MUŠEN); CTH 577 for mixed oracles (SU, KIN, and MUŠEN); CTH 579 for SU and MUŠEN, and CTH 580 for KIN and MUŠEN. Despite Laroche’s great achievement and the usefulness of his catalog, there are certain inconsistencies. The main one is the lack of distinction between MUŠEN oracles (i.e., the observation of free-flying birds) and MUŠEN ḪURRI (‘cave bird’ oracles, which appear to be a kind of bird extispicy). In addition, he classified the oracle texts partly by thematic criteria (CTH 561–569) and partly by technique (CTH 570–582). Later, researchers reorganized and refined the classification system. In addition, CTH 581, which Laroche had left unassigned, was allocated to the so-called ‘oracle letters,’ i.e., letters of various kinds referring to oracle procedures.

Despite all efforts, several overlaps and inconsistencies remain among the catalog numbers. A complete reorganization could create even more confusion, and the benefits of such a change would be uncertain. Therefore, Laroche’s classification has largely been kept in the “Konkordanz der hethitischen Keilschrifttafeln,’ created and maintained by S. Košak from 2002–2022 and continued after Košak’s death by Ch.W. Steitler, G.G.W. Müller, and J. Lorenz (hethiter.net/: hetkonk (2.plus)).

2.5 Further Key Publications

The first comprehensive studies of bird oracles appeared in 1973 and 1975. A. Ünal’s “Zum Status der Augures bei den Hethitern” (Ünal A. 1973a) treats the bird oracle extensively, discussing the augurs' role, terminology, and several key texts. Two years later, A. Archi published his article “L'ornitomanzia ittita,” a wide-ranging study covering flight terminology, the augurs' activities, technical vocabulary, bird names, and specific textual examples (Archi A. 1975e). Further important treatments include Beal R.H. 2002f, 65–71; Hazenbos J. 2007a; and Haas V. 2008a, 27–45. The first comprehensive philological analysis of all Hittite bird oracle texts, marking a milestone in the understanding of this oracle technique, was published by Y. Sakuma in his two-volume PhD thesis (Sakuma Y. 2009b). Its principal contributions include: the systematic derivation of grammatical and procedural rules governing bird observation protocols; the identification of the decisive role of the terms GUN-li- and taruyali-/tarwalli- and their combinations with SIG₅-za (‘from the favorable side’) and kuštayadi/kuštayaz (‘from the unfavorable side’) in determining the oracle outcome; the identification and zoological evaluation of more than thirty bird species names; and a refined CTH classification. A short summary of the main results has been provided by the same author in an article published in 2007 (Sakuma Y. 2007a).

In the framework of our research project, we have already published numerous new annotated digital editions of oracle reports and oracle letters. Unlike previous studies, editions of larger texts also provide a thorough examination of their historical and cultural contexts.

3. Findspots and Dates

Most Hittite bird oracle texts were discovered in the royal archives of the Hittite capital, especially in Temple I in the Lower City and several buildings on Büyükkale, but also in the Upper City and the ‘Haus am Hang’ (‘building at the slope’). As the archaeological context shows, many texts had already been discarded or used for secondary purposes, such as filling in walls, indicating that they were intended for immediate practical use rather than long-term preservation. Based on palaeographic criteria and, in some cases, content-related criteria, most of the texts can be dated to the Empire Period (second half of the 14^th to the end of the 13^th century), with the majority dating to the second half of the 13^th century. These documents are part of the extensive divinatory corpus from the final centuries of the Hittite kingdom, during which other oracular practices are also particularly well documented.

Bird oracle texts that have survived from the Middle Hittite Period show that ornithomancy was already well established by the late 15^th to the 14^th century BCE. However, in contrast to later documents, the reports from the Middle Hittite period are less standardized and show considerable differences in content and terminology compared to those of the Late Empire period.

Although the earliest reports of bird oracles date back to the Middle Hittite period, a narrative about Ḫantili I, who ruled in the mid-16^th century, indicates that ornitomancy was already being practiced during the Old Hittite Kingdom (KBo 3.63 obv. I 1–8; see Bawanypeck D. 2005a, 2; Hazenbos J. 2007a, 104; Haas V. 2008a, 28). The text notes that Ḫantili failed to conduct bird oracles during a campaign, which ultimately brought disaster upon the palace. It remains unclear, however, how the bird observation was conducted and how this practice resembled those of later periods. The same is true for the bird oracles mentioned in an inscription on a statue of King Idrimi of Alalaḫ (see Haas V. 2008a, 27 with note 99 and further literature).

4. Texttypes and Tablet Features

Overall, the texts can be divided into three categories: 1. Accounts of consultations conducted exclusively using bird oracles; 2. Accounts of investigations that combine bird oracles with other oracle techniques such as SU (examination of a sheep’s entrails performed by the diviner), KIN (a kind of symbol oracle practiced by the ‘Old woman’), and MUŠEN ḪURRI (examination of a bird’s entrails performed by the diviner); and 3. Accounts of bird oracles embedded within letters.

Most ‘pure’ bird oracle reports are subsumed under the catalogue number CTH 573, whereas reports of observations carried out by a combination of different oracle techniques are subsumed under CTH 577 (SU, KIN, MUŠEN, and MUŠEN ḪURRI); CTH 579 (SU, MUŠEN, and MUŠEN ḪURRI), CTH 580 (KIN, MUŠEN, and MUŠEN ḪURRI). Letters with embedded bird oracles are mainly subsumed under CTH 581. Further letters include CTH 186.1; CTH 188.1; CTH 188.2; CTH 190.1; CTH 190.2; and CTH 195.

‘Pure’ bird oracles are often inscribed on small, single-column tablets. Many of them exhibit poor-quality clay, hasty handwriting, and numerous erasures and errors, suggesting that they were produced during the oracle procedure. In contrast, larger tablets with more careful handwriting were likely created only after the event, based on the notes taken during the inquiry. This category also includes many oracle summaries that do not provide a detailed report of the oracle procedure and findings, as well as records referring to oracle procedures using different techniques (for oracle summaries versus comprehensive reports, see Hout Th.P.J. van den 1998c, 15–33).

5. Topics and Function

Most bird oracle texts that have come down to us concern the well-being of the Hittite king and the functioning of the state. Common topics include military decisions, health issues, royal succession, cultic issues, bad omens, and dreams. Like other oracle techniques, they function both as a diagnosis system (what is wrong? which deity is angry?) and as a form of negotiation with the gods (do you, O deity/gods, approve of this plan? Is the proposed offering or compensation appropriate?).

Insofar as the tablets are sufficiently well-preserved and contain information about the reason for the oracle consultation, we cannot see any discernible thematic preference among the various oracular methods.

The assumption that male oracle experts, such as augurs and diviners, primarily provided oracles concerning military campaigns and political issues like royal succession, while the ‘Old Woman,’ as the expert of the performance of KIN oracles and magical rituals, focused mainly on inquiries about illnesses and religious matters, is not supported by the available evidence. For instance, the comprehensive oracle text KUB 5.1+ (CTH 561.1), which was performed before a planned military campaign in Northern Anatolia, was primarily conducted using KIN procedures, with only a few SU oracles performed as a secondary method.

6. Experts and Institutional Setting

Late Hittite bird oracle texts regularly name the augur or a team of augurs responsible for the interpretation, and this naming is tightly integrated into the formal structure of the oracle paragraph: oracle question – requested result – observation report – verdict. The usual form for the latter is UM-MA PN ‘thus (speaks) PN’ followed by the oracle outcome: “it was confirmed/they (the birds) excluded (it).’ If the oracle followed another one in order to cross-check the result of the first one, the augur is mentioned with his title. In rare cases, the augur is mentioned only by his title: IŠ-TU (or TA) + (^LÚIGI.MUŠEN, ^LÚMUŠEN.DÙ or ^LÚIGI.DÙ).

Several indications from these dossiers suggest a collaborative practice and an internal hierarchy. The observation protocol is often introduced with phrases such as NI-MUR or aumen, meaning ‘we saw.’ Similarly, the verdict is frequently introduced with UM-MA PN Ù PN, which translates to ‘thus (speak) PN and PN.’ In some cases, even three augurs are named.

Ordinary experts are referred to by the titles ^LÚIGI.MUŠEN (lit. ‘bird observer’ or sim.), ^LÚIGI.DÙ (‘observer’), or ^LÚMUŠEN.DÙ (‘bird breeder, catcher’). Higher-ranking augurs hold the titles GAL ^LÚ.MEŠIGI.MUŠEN ‘chief of the augurs,’ UGULA ^LÚ.MEŠIGI.MUŠEN or UGULA ^LÚ.MEŠMUŠEN.DÙ, both translating to ‘overseer of the augurs.’ This internal hierarchy is also reflected in dossiers, according to which the observation was carried out by the junior augurs, whereas the final verdict was given by their overseer (cf., e.g., KUB 5.11 obv. I 14 (CTH 577); KUB 5.24 + KUB 18.57 obv. I 66–70 (CTH 577)). For more information, see Haas V. 2008a, 27–31 with further literature.

7. Technique: How a Bird Oracle Works

7.1 Question and expected outcome

A bird oracle typically starts with a specific question, distinguished by content (something desired or undesired) and grammar (positive or negative phrasing). This leads to four patterns:

Positive content, positive grammar: ‘let the birds confirm (the desired)’ (e.g., ‘let the birds confirm our success’).

Positive content, negative grammar: ‘let the birds exclude (the undesired)’ (e.g., ‘let the birds exclude that we are not successful’).

Negative content, positive grammar: ‘let the birds confirm (the undesired)’ (e.g., ‘let the birds confirm our defeat in battle’).

Negative content, negative grammar: ‘let the birds exclude (the undesired)’ (e.g., ‘let the birds exclude that we will not be defeated’).

A second observation often follows as a countercheck, using the opposite formula to verify or falsify the first result.

If the countercheck is performed with a different technique, it is introduced by the phrase IŠ-TU ... ER-TUM QA-TAM-MA-pát, which means ‘through (the diviner/Old Woman/ENSI-priestess) the very same question’ or IŠ-TU ... KI.MIN, meaning ‘through (...) ditto.’

If the countercheck is carried out by a second bird observation, it is introduced by the phrase IGI-anda (menaḫḫanda) ... MUŠEN^ḫi.a NI-MUR or IGI-anda SAG.KI-anza DAB-annaš/DAB-anna (ḫanza appannaš/appanna) ‘as a countercheck of confirmation/to confirm we saw.’ For a detailed overview with further literature, see Sakuma Y 2009b, I, 23–28.

7.2 Structure of the Report

A complete New Hittite bird oracle inquiry typically contains:

Question and expected outcome: ‘let (the birds) confirm (it)’ (ḫandandu) or ‘let (the birds) exclude it’ (peššiyandu).

One or more observation protocols (often three parts per protocol), each describing which bird(s) entered the field, from where they came, and where they flew away. Sometimes additional behaviors, such as beak movements, calling, perching, fighting, or dropping, are recorded.

Name and title of the author(s) and their final statement: ‘it was confirmed’ (SI×SÁ-at) or “they (i.e., the birds) excluded it’ (peššer). For a detailed overview, see Sakuma Y. 2009b, I, 23–38).

7.3 The Observation Field and the Birds’ Movements

The standard descriptions of the Hittite bird oracle reports indicate that the augurs observed the flight patterns and other behavior of the birds within a defined observation area rather than just the open sky. The texts consistently mention that a bird came from a specific direction and departed in a particular direction. They also refer to local reference points, such as a river in Middle Hittite texts and a road in New Hittite texts. Furthermore, they contain a rich inventory of local expressions, such as ‘behind’ (EGIR = appan), ‘in front’ (peran, abbr. pé-an), ‘in the middle’ (2-an = takšan), ‘from the favorable side’ (SIG₅‑za), ‘from kušta(i-), i.e. the unfavorable side’ (luw. kuštayadi, Hittite abbr: ku-uš-za, ku-za, ku-uš, ku), ‘up’ (UGU =šarā) and ‘down’ (GAM = katta). Additionally, the texts use terms whose exact meanings remain unclear, but that seem to refer to the birds’ localization, flight paths, and, occasionally, the orientation of their beaks as well. Among them are zilawan (abbr. zi, zi‑an), tentatively translated ‘lengthways, straight ahead (to the end),’ and pariyawan (abbreviated pa‑ri, pa‑an), which likely denotes a slanting or diagonal orientation, as well as the oppositional pair GUN‑li‑ and tar(uya)li‑. These last two accompany bird names either as adverbial adjectives when the bird is the subject of a clause (e.g. ‘a ḫarrani‑ bird came/flew (as a) GUN‑liš/tar(uya)liš (bird)’), or as adverbs in clauses with one or more bird names in the accusative and the verb ‘to see’ (NIMUR: ‘we saw a ḫarrani‑ bird in GUN‑li‑/ tar(uya)li‑ manner/orientation’).

Abbreviations of the adjectival forms of GUN‑li‑ include GUN, and more rarely GÚ, while adverbial GUN‑li‑ forms are abbreviated as GUN‑li₁₂‑an or simply GUN‑an. For tar(uya)li‑, the abbreviations tar‑liš, tar‑u and tar‑u‑iš stand for adjectival forms, and tar-li₁₂-an, tar‑u and tar‑li₁₂ for adverbial forms. This morphology already hints that GUN‑li‑ and tar(uya)li‑ describe not only abstract areas of a field but characteristic orientations or modes of flight that can be predicated either of the bird or of the act of seeing.

Although the general meaning of most of these terms is more or less clear, it remains difficult to determine their relation to specific parts of the observation field. Where is the ‘favorable side’ (SIG₅) and where is its opposite, the kušta(i)- or unfavorable side? Why is the latter not denoted as NU.SIG₅? Where exactly is the road, and where is the ‘middle’? The problem becomes even more acute when several locative expressions are combined in one clause, such as ‘behind the road,’ ‘in front from the favorable side,’ or ‘it (i.e., the … bird) came up in the back from the favorable side, towards the sun.’ To understand such complex descriptions, it is, inter alia, important to determine the augur’s position relative to the field. If the augurs stood on the road, as R. Beal has suggested, they would be positioned between two fields, one in front and one behind them (Beal R.H. 2002f, 66). They would first observe birds in the field in front of them, then turn around to observe the field described as ‘behind the road’ (EGIR KASKAL), which initially lay behind them. Alternatively, following Y. Sakuma, one may assume that both fields lay in front of the augurs from the outset and were intersected by the road (Sakuma Y. 2009b, I, 86–89). In that case, the field ‘behind the road’ would denote the more distant zone, although the texts do not define the geometry with complete clarity.

7.4 Patterns of Combination and Sakuma’s Quadrant Model

Starting from this network of locative expressions, the next step is to examine the combinations in which they occur. Based on previous research by Archi A. 1975e and Beal R.H. 2002f, 65–71, Sakuma systematically examined the actual combinations in which they occur (Sakuma Y. 2009b, I 81–130; Sakuma Y. 2013a). His key insight is that the technical terms do not combine freely, but appear in highly restricted patterns, especially in the clauses describing from where a bird came flying (uwa‑sentences) and to where it flew away (pai‑ sentences). These patterns can be grouped into a small number of formula types, which in turn can be mapped onto a four‑part field. A central role is played by GUN‑li‑ and tar(uya)li‑. In sentences with the verb NIMUR (auwen) ‘we saw,’ they are attested only with the following combinations:

GUN‑li‑ with EGIR UGU SIG₅‑za ‘at the back, up, from the favorable (side)’; peran kuštayati/ku-za ‘in front, from the kušta(i)‑ side’ (i.e., the unfavorable side)

tar(uya)li‑ with EGIR GAM kuštayati /ku-za ‘at the back, down, from the kušta(i)‑ (side)’; peran SIG₅‑za ‘in front, from the favorable (side)’

From this distribution, Sakuma concludes that GUN‑li‑ and tar(uya)li‑ stand in opposition and each designates sightings of birds in specific quadrants that lie diagonally opposite each other. At the back, GUN‑li‑ marks the quadrant on the favorable side, which he defines as the right-hand side; in front, GUN‑li‑ marks the quadrant on the kušta(i)- (or unfavorable) side, which he believes to be the left‑hand side. For tar(uya)li‑ the pattern is reversed: in the back, tar(uya)li‑ is the quadrant on unfavorable (or left-hand side), whereas in the front, it is the quadrant on the favorable (or right-hand) side. Together with the explicit references to ‘front’/‘back’ and to favorable vs. unfavorable (kušta(i)‑ side), this yields a four‑quadrant field divided horizontally into a front and back half, and vertically into a right‑hand favorable and a left‑hand unfavorable zone.

The attestations of pariyawan (pa‑an) and zilawan (zi‑an) also match this reconstruction. pariyawan occurs in uwa‑ sentences only together with GUN‑li‑ or tar(uya)li‑, never with SIG₅‑za or kuštayadi, which suggests that it denotes a slanting or diagonal course across the field rather than a move simply from the favorable or unfavorable side. The same term is used for the position of a perching bird and its beak; given the limited repertoire of possible body and beak postures, it is plausible that a ‘slanting’ position attracted attention more readily than a straight one, so that the term came to mean ‘diagonal, oblique’ both for flight paths and body and beak positions. In pai‑ sentences, the same restriction to GUN‑li‑ and tar(uya)li‑ holds; here pa(riyawa)n apparently again describes an oblique movement with which the bird leaves the field.

The term zilawan (zi‑an), which is also unattested outside the bird oracle corpus, is related to the Luwian adverb zila- ‘after, afterwards, then’ and the noun ‘end, result’ (see Rieken E. 2019b with further literature). As a term describing the flight of a bird, it might best be translated as ‘lengthways, straight ahead (up to the end).’ This meaning is supported by passages like KUB 18.12+ obv. 20–21, where two eagles are said to come ‘at the back, up, from the favorable side’; one then flies zi‑an, while the other bird ‘turns back’ (EGIR-pa dāš).

When all locative expressions are taken together, and other information is considered – such as the widespread idea that the right side is favorable and the left side unfavorable –Sakuma’s model appears broadly convincing. He posits an observation field consisting of two identical four‑quadrant fields, one ‘in front of’ and one ‘behind’ the road. In practical terms, the augurs’ physical location is not essential: whether they stand on the road and turn around, or face both fields at once, the same pattern of quadrants relative to their body and to the favorable/right vs. kušta(i)-/left side emerges. Crucially, Sakuma shows that the ultimate outcome of an oracle correlates strongly with whether the birds have flown predominantly from quadrants associated with the favorable or the unfavorable side. This suggests that the elaborate system of orientations converges, at least in the written reports, on a primary evaluative axis: ‘proper, favorable’ vs. ‘deviating, unfavorable.’

7.5 The Alleged Sketch of the Field

A key piece of evidence for Sakuma’s reconstruction of the observation field is a fragmentary drawing found on a cuneiform tablet, published as KUB 49.60 (CTH 577). The drawing depicts the remains of the right side of a field. By restoring the left side, it appears to form a square composed of four equal squares, divided by two parallel diagonal lines that cross each square and converge in the center. Along with the textual evidence, Sakuma interpreted the right side of the field as the ‘favorable side,’ divided into two quadrants, with the rear quadrant as the GUN-li- area and the front quadrant as the tar(uya)li‑ area. Conversely, the left side would then correspond to the unfavorable (kušta(i)- area), also divided into two quadrants, with the rear quadrant to be identified as the tar(uya)li‑ area, and the front as the GUN-li- area (Sakuma Y. 2009b, 84–86; 105; II, 360; Sakuma Y. 2013a, 220–223). Based on suggestions by Volkert Haas and Joost Hazenbos, Sakuma proposed reading the sign traces below the middle line of the field as tar-li (according to a verbal communication by Hazenbos) and the sign traces in the next line as pé SIG₅-za (as per Haas V. 2008a, 33). Both readings are, however, highly questionable, which Sakuma indicates by using question marks and exclamation marks following the alleged pé (rendered by him as pé.^!?) and SIG₅-za (rendered as ⸢SIG₅^!?-za⸣). Regarding the suggested reading tar-li, Sakuma correctly notes that this writing is not attested in New Hittite bird oracles, where it is consistently spelled with the sign LIŠ (tar-li₁₂). A more plausible reading appears to be NU.ŠE-du, which Sakuma also considers as an alternative (Sakuma Y. 2009b, 360). However, this interpretation presents issues when viewed as part of a legend associated with the observation field of bird oracles. This is because augurs typically ask the birds to either confirm (MUŠEN^ḪI.A SI×SÁ-an-du) or exclude the question (MUŠEN^ḪI.A peššiyandu). In contrast, the phrase ‘let (the oracle) be unfavorable’ is typical of KIN and SU oracles. Moreover, the alleged pé before the alleged SIG₅-za is rather an -i.

Another problem is that some of the signs are not placed within or next to the sketch, but instead appear below the vertical line marking the left side of the intercolumnium or the lines of the supposed observation field. This positioning makes it questionable to interpret them as labels for specific areas of the field. It even remains unclear whether the sketch depicts the observation field of a bird oracle. If the reading NU.SE-du is correct, the markings may be remnants of a KIN or SU oracle that have been partially erased and overwritten by the intercolumnium and some lines related to the sketch. The situation becomes even more complex when we consider that a similar sketch is found on the reverse side of the tablet KBo 41.141. In this case, it is unlikely that the drawing represents an observation field for bird oracles, as the oracles recorded on the tablet are SU oracles, not bird oracles (pace Hazenbos J. 2007a, 97 with note 10; Haas V. 2008a, 33 with note 128). In conclusion, the meaning of the technical terms in the bird oracles must primarily be determined based on textual evidence.

7.6 How the Verdict is Derived

From fully preserved examples of the New Hittite period, it becomes clear that the augur evaluates the whole sequence statistically along the axis favorable vs. unfavorable: if the majority of decisive flights in a protocol come from the favorable side (SIG₅-za), the issue addressed in the oracle question was considered to be confirmed; if the birds predominantly come from the unfavorable side (kušta(i)- side), the issue was considered to be excluded by the birds.

In Middle Hittite texts, the fundamental principle of counting weighted favorable and unfavorable movements appears already to be established, though the process was less structured compared to that in the Late Empire period.

8. Bird Names

The class term for ‘bird,’ wattai- gen. com., is itself rarely used in the oracles. Furthermore, the majority of bird names occur without the determinative (MUŠEN). Their meanings remain mostly unclear. Among the birds that can be identified are the eagle (TI₈^MUŠEN, ḫaran-); the raven (UGA); and the falcon (SUR₁₄.DÙ.A, kallikalli-). Names of unknown meaning for birds and other flying animals, such as bats, are: ālliya-, alili-, aram(n)ant-, arši(n)tatḫi-, ḫalliya- (a kind of bat?), ḫalwašši-, ḫarrani-, ḫarliti- (hapax in KUB 18.12 obv. 10), ḫaššiḫarši- (hapax in KBo 32.123 obv.? 7´), ḫastapi-, ḫu(wa)ranni-, ḫūša-, ḫuwa(ra)-, kalmuši-, kaltarši-, kani/ešdu- (a certain kind of hawk/or bird of prey), laḫḫanzan- (a certain kind of water bird; water fowl? duck?), marašši- (hapax in KuT 50 obv. 13), maršanašši- (hapax in HKM 47, rev. 54), pattarpalḫi- ‘wide-wing’ bird (a certain kind of hawk or bird of prey), šaluwa(ya)-, šaluwašalwa- (hapax in KUB 22.3 rev. 4), šalwinieš-, šulupi-, šūrašūra (crow or raven?), tapašši-, uwaranni- (hapax in KBo 47.225 obv.? 4), ura/iyanni- (woodpecker?), wešši-(hapax in HKM 47 obv. 24), zamnašši- and the logographically written birds KA₅.A (‘fox-bird,’ perhaps megabat, PÉŠ ‘mouse-bird,’ possibly also a bat, UR.BAR.RA (‘wolf,’ DAAM 1.21 obv. 9, 13), and MUŠ ‘snake’ (hapax in Or 90/335 obv. 12´). Uncertain are ḪAR.ḪAR (KBo 32.123, MH), and MUŠEN.MUŠ ‘snake-bird’ (KBo 47.63 r. c. 5´).

9. Middle Hittite versus Late Hittite Texts

Middle Hittite bird oracles occupy an important transitional position. A major distinction to the Late Hittite reports is that in the Middle Hittite texts, the movements of the birds are often described in relation to a river that the birds are frequently said to cross. A further difference is that New Hittite reports normally present a full sequence of oracle question, requested result, observation protocol, augur name and/or title, and achieved result, whereas Middle Hittite texts never preserve the entire scheme and can be reduced to little more than the observation itself. The oracle questions are broadly comparable, but Middle Hittite texts usually omit both expected and achieved oracle results, which survive only in a few short formulas. Augur names and titles are also rarely given in Middle Hittite texts. If so, they are not mentioned within the protocol itself. The vocabulary of bird names is largely shared; however, Middle Hittite texts often show more phonetic spellings, for example, ḫara(n)-^MUŠEN instead of TI₈^MUŠEN ‘eagle,’ and frequently attach the determinative MUŠEN to syllabically written bird names, which is uncommon in New Hittite texts. Additionally, technical terms in Middle Hittite texts are often given in full Hittite phonetic spelling, whereas New Hittite texts frequently feature logographic or heterographic spellings and abbreviations.

Key examples of Middle Hittite bird oracles include:

1) KBo 66.70 + KUB 18.5 (CTH 573.113, Temple 1 of Ḫattuša/Boğazköy), which describes almost all bird movements in relation to the river; 2) HKM 47 (CTH 581.32, Tapikka/Maşat), which is an almost fully preserved letter containing an unusually detailed, three-day bird oracle report with several distinct sections and rare bird names and phrases; 3) The letters KuT 49–50 (CTH 190.1 and CTH 190.2, Šarissa/Kusaklı), which both feature notable oracle terms and provide important information about the use of bird oracles in specific situations and their relation to other forms of communication between gods and humans.For a detailed overview of the Middle Hittite reports, see Sakuma Y. 2009b, 252–333 (with further literature).

10. Open Questions

While Sakuma’s quadrant model explains much of the data and provides a valuable analytical framework, several open questions remain. One difficulty is the apparent mismatch between the richness of the described movements and the relatively simple evaluative rule. The protocols painstakingly distinguish between straight and diagonal flight, between front and back, up and down, GUN‑li‑ and tar(uya)li‑. Yet, according to what we can say from the better-preserved texts, the decisive factor for the outcome appears to be essentially the tally of flights from SIG₅‑ vs. kušta(i)- areas. It is hard to believe that the other parameters were entirely devoid of interpretative significance. Unfortunately, due to the lack of evidence, we cannot gain further insights into the significance of these other parameters.

Finally, the adjective iparwašši‑ ‘western, oriented to the west,’ especially when appearing together with explicit references to the sun, indicates that the observation field is not the only reference system at play. iparwašši‑ is commonly translated as ‘in the west,’ but as an adjective accompanying the name of a bird that is said to be seen in the sun or flying ‘towards the sun’ indicates that there were two overlapping frames—an egocentric frame (right/left and front/back) and a cosmic frame (orientation toward the setting sun). Here, too, Sakuma’s model helps by highlighting regularities, but it may not capture the full complexity of the augurs’ conceptual space.

11. List of terms and translations