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

‘Cave bird’ Oracle (MUŠEN ḪURRI)

1. Short description

Among the various oracular techniques used by the Hittites, the MUŠEN ḪURRI oracle holds a unique and somewhat puzzling position. Unlike augury (MUŠEN), which involves observing a large number of various kinds of free-flying birds in a defined area, this oracle technique is only performed with one specific bird named MUŠEN ḪURRI. The exact nature of this bird remains unclear, but the term ḪURRI, which derives from Akkadian ḫurru ‘hole, cavity, cave,’ indicates that it is a bird nesting in burrows or caves. Based on this, researchers have suggested that it might be a rock partridge, a ruddy shelduck, or a similar waterbird (for a detailed overview, see Tognon R. 2005a, 12–15).

In contrast to the observation of free-flying birds, the MUŠEN ḪURRI oracle was not performed by the augur or bird-watcher (^LÚIGI.MUŠEN, ^LÚIGI.DÙ, ^LÚMUŠEN.DÙ). Instead, it was conducted by the ‘diviner’ (^LÚḪAL or ^LÚAZU, Akkadian bārû), who also performed sheep extispicy (i.e., the observation of the liver, gallbladder and intestines of a slaughtered sheep), and the šašta- or ‘bed’ oracle (i.e., the observation of the behavior of two rams prior to extispicy). Another difference between augury and the MUŠEN ḪURRI oracle is that the latter expresses its results in the binary terms ‘favorable’ (SIG₅) and ‘unfavorable’ (NU.SIG₅), similar to extispicy. By contrast, in augury, the requested outcome is either ‘confirmed’ by the birds (SI×SÁ-at = ḫandaittat) or ‘excluded’ (arḫa peššer, meaning ‘they (the birds) excluded it’). Another similarity with sheep extispicy is the common request for a two-fold outcome: the first observation of the animal (or its entrails) is requested to be favorable, while the second is requested to be unfavorable (or vice versa). The standard formula is ‘let the first ḪURRI bird be favorable and the second one unfavorable.’

2. Research history

The history of research on the MUŠEN ḪURRI oracle can be divided into three main phases. Until the 1970s, scholars primarily focused on publishing handcopies, along with sporadic translations and transliterations of individual passages and references. The first handcopies were published in 1916 by H.H. Figulla in KBo 2 (as KBo 2.2 and KBo 2.6), and later by A. Walther in KUB 5, KUB 6, KUB 16, KUB 18, and KUB 22, which were released between 1922 and 1928.

KBo 2.2 was then made accessible to a broader audience with a German translation by J. Friedrich in 1925. In this translation, Friedrich translated the terms for the various oracle techniques documented in the text – SU, KIN, and MUŠEN ḪURRI – as “die Fleischvorzeichen” (‘flesh oracle/omen’ = SUᴹᴱ^Š), “die Nachricht” (‘message’ = KIN), and “die Vogelzeichen” (“bird omen” = MUŠEN ḪURRI). He notes the following regarding the latter method in a footnote (Friedrich J. 1925a, II, 23 note 9): “Wörtlich: ‘Dann soll das Steinhuhn ungünstig sein’. Dieser Vogel wird von den Hethitern zum Orakel benutzt.” These remarks suggest that the interpretation of MUŠEN ḪURRI as a ‘rock partridge’ (‘Steinhuhn’) was clear and uncontested. However, this was by no means the case. For many years, researchers debated whether the bird – which is referred to in Akkadian texts as iṣṣur ḫurri – was a falcon, an owl, a rock partridge, or a species of shelduck (tadorna). For an overview, see Tognon R. 2005a, 12–15.

A new phase of research started in the 1970s with the accelerating publication of handcopies of new text finds, critical editions and systematic treatments of oracle texts in general. In a foundational study on the Hittite bird oracles (MUŠEN), A. Archi also addressed the nature of the ḪURRI bird procedure, proposing that it was a form of extispicy adopted by the Hittites from the Hurrians (Archi A. 1975a, 140).

The third phase began at the beginning of the 21^th century and continues to this day. It is characterized by a comprehensive examination of the entire corpus of the texts, their historical, cultural, and linguistic analysis, and the goal of making these sources accessible to a wider audience. Notable overviews are Hout Th.P.J. van den 2003e; Beal R.H. 2002f, 71–72, and Haas V. 2008a, 55. A significant contribution to the field is R. Tognon’s unpublished dissertation, which presents the first complete edition of all MUŠEN ḪURRI oracles categorized under CTH 574, as well as mixed-oracle documents that incorporate the MUŠEN ḪURRI technique (Tognon R. 2005a). Additionally, the same scholar provided an edition of the oracle inquiry KUB 5.7, which was conducted using the ḪURRI bird technique (Tognon R. 2004a).

3. Date and Findspots

So far, no MUŠEN ḪURRI oracle text has been discovered that can be dated to the period before the 13^th century BCE based on palaeographic criteria. However, since oracle texts were typically not archived for long periods, one should not read too much into their absence. In contrast, the use of the ḪURRI bird in purification rituals borrowed from the Hurrians during the Middle Hittite period makes it quite likely that the Hittites already employed this bird for divination purposes prior to the 13^th century (for attestations of the ḪURRI bird aside from oracle reports, see section 5).

Most tablets documenting oracle inquiries performed with the MUŠEN ḪURRI technique come from the Hittite capital Ḫattuša. The main findspots are Büyükkale Buildings E, C, and D, as well as Temple 1 in the Lower City and the House at the Slope (Haus am Hang). It appears that many oracle tablets had already been discarded by the Hittites themselves and were later used as fill material in the citadel walls and building foundations. For a more detailed overview, see Tognon R. 2005a, 28–29.

One MUŠEN ḪURRI text (AT 454) was found in Alalaḫ (Tell Atchana in northern Syria). More recently, oracle reports including MUŠEN ḪURRI oracle procedures have come to light at Kayalıpınar (ancient Šamuḫa), and were edited by Miller J.L. 2019a as DAAM 1.56; DAAM 1.57; DAAM 1.58; and DAAM 1.59.

4. Topics

The thematic range of MUŠEN ḪURRI oracles is narrower than that of other Hittite oracular techniques. While virtually every aspect of life – military campaigns, the health of the king and queen, accession to the throne, political intrigue – could be investigated through oracles of various kinds, the MUŠEN ḪURRI technique is (almost) exclusively concerned with religious matters, specifically identifying omissions and transgressions in the cults of various deities. The typical situation presupposed by a MUŠEN ḪURRI oracle is that a deity has become angry because offerings were not given, festivals were not celebrated, or the purity of a temple was not maintained. The oracle then identifies which deity is wrathful and why. Recurring cultic topics include omissions of festivals; omissions of daily bread offerings; profanation of temples by the entry of impure humans or animals; neglect or disappearance of cult objects (for a detailed overview, see Tognon R. 2005a, 23–26).

The only potential exception to the exclusively cultic character of MUŠEN ḪURRI topics is ABoT 16 (CTH 574), which mentions the lands of Alziya and Iššuwa. According to Beal R.H. 2002f, 81 with note 149, military matters, specifically the deportation of defeated enemies, may be involved. However, due to its fragmentary state of preservation, the content remains unclear (see also Tognon R. 2005a, 23–26).

5. Experts and technique

The exact nature of the MUŠEN ḪURRI procedure has been a topic of debate, primarily because reports seldom describe the physical procedure, usually only presenting the binary outcome (SIG₅ or NU.SIG₅).

In a foundational study, A. Archi addressed the nature of the ḪURRI bird procedure, arguing that the technique represented a form of extispicy (Archi A. 1975a, 140). The basis for his assumption was that the oracle was performed by the diviner (^LÚḪAL) who also conducted sheep extispicy. Furthermore, he pointed out that KUB 18.14 rev. III 12–15 contains a description of the oracle’s findings. Thus, rev. III 12 runs as follows: nu MUŠEN ḪUR-RI NU.SIG₅-du e-ra-i-iš DIB-an-za NU.S[IG₅] (“Let the ḪURRI bird be unfavorable. eraiš is taken. Unfa[vorable].” A similar passage appears at the end of the following paragraph (lines 14–15): “Let the ḪURRI bird be unfavorable. The zizzipki- is on the right. Unfavorable.”

As an even more striking example, Archi cites KUB 18.27 line 10, where the term ḫatḫiš appears. Since this term likely relates to ḫataḫi-, which is attested in sheep extispicy (KBo 10.7 obv. II 32; rev. III 10), Archi reasoned that the Hittites adopted the ḪURRI bird oracle from the Hurrians. (Archi A. 1975a, 140).

However, Archi’s extispicy hypothesis has been challenged. As R. Beal noted (Beal R.H. 2002f, 72, note 101), the fragmentary passage containing ḫatḫiš likely forms part of a description of a SU oracle rather than a ḪURRI bird oracle. Therefore, Beal suggested as an alternative scenario that the procedure might have involved a diviner reaching into the ḪURRI bird’s nest and determining the outcome based on whether he grabbed the bird’s tail, which would be considered favorable, or its head, which would lead to being bitten and deemed unfavorable. In this interpretation, the opaque terms erai- and zizzipki- could, for instance, refer to rare external markings on the bird that would influence the outcome. As a further possibility, Beal R.H. 2002f, 72 suggested that the MUŠEN ḪURRI was a poison oracle similar to the chicken oracle of the Azande of Sudan, in which the bird either survives or dies after ingesting a substance, and unusual marks (erai-, zizzipki-) discovered post-mortem would modify the result.

While the head or tail scenario seems unlikely, Beal rightly points out that the procedure may have also involved observing external features of the bird, such as feather color and specific deformities in body parts. This possibility is further supported by evidence of similar practices in Syria and Mesopotamia (see De Zorzi N. 2009a, 87).

Recently, however, Sakuma Y. 2025a has provided additional arguments supporting the extispicy theory. The most significant evidence comes from KBo 46.87 (+) KBo 59.84 (Sakuma Y. 2025a, 215–216). The relevant passage runs: IGI-an-da SAG.KI-[an-za DAB-an]-na-aš SUᴹᴱ^Š ER-u-en nu MUŠEN ḪUR SIG₅-ru SIG₅ “As a countercheck for confirmation, we inspected the exta: let the ḪURRI bird be favorable. Favorable”. This passage indeed suggests that the MUŠEN ḪURRI oracle is not only a form of extispicy but is also referred to by the same term used for sheep extispicy, namely SUᴹᴱ^Š, meaning ‘flesh (signs).’ However, since the term SUᴹᴱ^Š is otherwise exclusively used as a designation of the sheep’s exta, it is not unlikely that the unusual phrasing may result from a scribal error – a possibility that Sakuma does not consider. Instead, he reinforces his hypothesis by pointing out that counterchecks of oracles conducted with the same technique are usually introduced by the formula IGI-anda SAG.KI-(an)za DAB-a[nnaš (menaḫḫanda ḫanza appannaš), meaning ‘as a countercheck for confirmation,’ or by its opposite NU.SIG₅-anni (kallaranni) ‘as a countercheck for misfortune.’ In contrast, counterchecks using another oracle technique are typically introduced with the formula IŠ-TU / TA ... (title of oracle expert) ER-TUM QA-TAM-MA-pat / KI.MIN. Based on this, Sakuma concludes that both techniques were classified as the same type of divination. However, this argument requires nuance. First, in KpT 1.58+KpT 1.61, the ḪURRI bird oracle is not immediately followed by a SU oracle; a cross-check through the ENSI woman intervenes. Second, the formulaic choice between IŠ-TU / TA ... and menaḫḫanda ḫanza appannaš rests mainly on whether the same or a different ritual expert performs the subsequent check. Since both methods are within the diviner’s domain, transitions between them are naturally marked by menaḫḫanda ḫanza appannaš. The evidence presented by Sakuma can therefore not be considered definitive proof of the extispicy hypothesis.

Beyond divination, the ḪURRI bird also appears in purification rituals. In the Middle Hittite itkalzi-ritual (KBo 20.129 obv. I 13–16), the diviner (^LÚAZU) swings the bird around a divine statue and the ritual patron; in the birth ritual ABoT 17+, the patili- priest swings it around the furniture of a birthing chamber to drive away evil influences; and in the Papanekri ritual (KBo 5.1 obv. II 55–57), the patili- priest swings a ḪURRI bird around the deities and the ritual utensils as a purificatory act. The use of the same bird for both divinatory and apotropaic purposes highlights its specific role in Hittite religion: the MUŠEN ḪURRI was associated with the diagnosis of divine anger, the removal of impurity, and the restoration of divine favor. For further attestations of the bird outside the divinatory texts, see Tognon R. 2005a, 13; for an attestation in the cult inventory KUB 38.38 obv. 8′, mentioning two golden ḪURRI birds as parts of a divine statue, see Burgin J. 2022b, 490–491.

6. Combination with Other Techniques

The ḪURRI bird oracle rarely stands alone. Even many of the texts classified under CTH 574 as ‘pure’ ḪURRI bird oracles might originally have belonged to larger investigative sessions (see also Tognon 2004–2005, 29). Furthermore, some reports of oracle investigations with two or three techniques show that the ḪURRI bird oracle was frequently used to negotiate with the deity about the nature and quantity of offerings, while other oracle techniques assess higher-level issues, such as the deity’s anger and reasons for it. A key example of this interplay is the oracle investigation recorded in KUB 22.70 (CTH 566), which uses both sheep extispicy and ḪURRI bird oracles (B. Christiansen (ed.), hethiter.net/: CTH 566). A plausible explanation for this division is that the MUŠEN ḪURRI oracle was a relatively quick and less expensive procedure, well-suited to answering the many specific follow-up questions that a cultic investigation generated.