Close Encounters of the Feathered Kind: Orpheus and the Birds

Abstract

Birds were observed in divinatory rituals in antiquity. This was the most significant process that regularly involved looking closely at birds. At the same time, birds were intimately connected to human perceptions of the natural world and, through their capacity to ascend skywards, with the supernatural world. By studying two neighbouring areas connected by the myth of Orpheus and divinatory birds, we can begin to appreciate the role of birds (and bird-like beings) in the culture of Macedonia and Thrace. Birds played a central role in prophecy (ornithomancy). Literary, archaeological and zooarchaeological data from the 6th to 4th centuries BCE provide the means of uniting different types of evidence. The literary and archaeological evidence provides a broad perspective for understanding the still-limited zooarchaeological data. Birds of prey were among the key divinatory creatures regularly observed, while cranes and a variety of lacustrine and meadow birds were among those most regularly observed and hunted. Winged creatures (human–animal hybrids) form some of the most important creations of the human imagination in southern Europe, with distinctive local variants in Macedonia and Thrace.

1. Introduction

Why Orpheus and birds? Most ancient literary and artistic representations of the mythical Orpheus associated him with the charming of animals, but some of the earliest surviving accounts connect him specifically with birds. This early association also points to a motive for this choice, namely the link between Orpheus as a seer, as well as a singer, on the one hand, and divination on the other, a practice in which bird behaviour played a key role. Early mythical evocations of Orpheus connected him with the northern Aegean and the Bosporan Straits, which featured in the stories of the Argonauts. These stories make the coastal regions of the north Aegean that correspond to ancient Macedonia and Thrace a good place to start an investigation of how and why ‘Orpheus’ (however we understand this mythical person) and birds became associated in literature and in art. Geographically, the area in question corresponds to the lowlands around the Thermaic Gulf in ancient Macedonia, and Thrace south of the Balkan range. It is in this area that legends about Orpheus were notionally located.

My curiosity about Orpheus and birds was aroused by an item from my own excavations at Pistiros, in Thrace: the top of a painted Attic askos, or oil flask, showing a siren, in the form of a woman’s bust attached to a bird’s body, with upswept wings (Figure 1). To begin with, there seemed no connection at all to Orpheus. Sirens appear occasionally on imported Attic painted pottery, in various parts of the Mediterranean. Connections between this image, Orpheus and prophetic behaviour only emerged later.

To understand these items fully, we need to explore them in a broader cultural perspective, as well as in their place of origin, that is, in Thrace. It is therefore useful to compare them with two thematically related but geographically remote items: a bronze askos, also in the form of a siren (ca. 470–460 BCE, see Tsiafakis 2001; formerly in Malibu), and a large-scale terracotta statue group, formerly in the Getty Villa, representing a musician, in the style of Orpheus, with two sirens, recently published by Maria Lucia Ferruzza (ca. 330–300 BCE, Ferruzza (2016), pp. 9–24, Cat. Nos. 1–3; Figure 2).1 Both the bronze askos and the statue group belong to the artistic repertoire of southern Italy, far away from the probable originating centres of the stories with which they are here associated. However, together these two examples represent the polarities that the subject matter of this paper incorporates. On the one hand, we see the strange world of half-human ‘demons’, and, on the other, the sophistication of culture and song. Birds are the link and ultimately the explanation for the figures represented: the bird-bodied siren flask, and the bird-legged sirens in the statue group, whose behaviour is modified by the song of the seated musician, whether Orpheus, or, more likely (following Ferruzza), the deceased who was commemorated in the style of this very mythical singer. In case we should feel that the monstrous bird of the askos is something quite alien to the world of the philosophers, we have only to look at the sirens of the statue group to appreciate that the monstrous and the terrifying occupy the same universe as the calm and the philosophical, represented by the seated musician. The connection between rational (philosophical) and irrational ideas (in philosophy, but also in practical life) is underscored further in the remarkable finds from Tomb A at Derveni, Macedonia, where a papyrus text with ’Orphic’ writings was found alongside a range of assorted grave goods, including a heap of astragali. At the same time, the artefacts from Magna Graecia encapsulate ideas that apparently originated outside Italy, in the northern Aegean in the case of the statue group, and perhaps in Egypt, in the case of the siren-shaped askos, since this image bears a strong resemblance to the Egyptian ‘Ba’ bird (discussed further below).2

The mutability of ideas and images across space and time complicates the ways in which we conceive specific historical cultures and the objects that emerged within them. This paper will delve into the problem of how we can begin to prise apart the stories behind bird hybrids. My first aim in this paper is to focus on the tale of Orpheus and birds. This will enable connections to be made between the story, the associated images, and one geographical and cultural context in which we can view such a story in operation. A second aim of this paper is to consider how ideas behind images of bird hybrids were disseminated across Eurasia in the 1st millennium BCE. The interconnection of ideas that I will be following below shows that some of the elements contributing to the concept of ‘Orphism’ can be seen at work in the practices of divination, and of prophecy more broadly, in which birds had a role to play. At first sight, the practices around divination do not help us to understand the bird-like beings that accompany the singer of the Getty statue group (Figure 2) and the bird-bodied siren askos (Figure 1). Some of the problems that arise in our attempts to interpret bird-like images are connected to the compartmentalisation of knowledge and areas of study about antiquity. Creatures that we can identify in the natural world are studied as early representations of known animal species, while the strange hybrids that occur even more frequently as artistic images than ‘natural’ creatures are consigned to the dubious category of ‘monsters’ (LIMC Suppl. 1 (2009) 143–56 sv. Daemones; pp. 359–71 sv. Monstra).

2. Methods and Materials

Representations of imaginary, bird-like creatures suggest that birds had a special significance within the animal kingdom in remote times. Winged beings are among the commonest representations of zoomorphic forms in Mediterranean and Eurasian areas. The importance of birds in antiquity is hard to grasp, despite their manifest significance in divinatory rituals. There are few surviving texts that describe how the movement and passage of birds was observed for the purpose of divination, but there are traces of standard procedures (such as the 6th century BCE inscription from Ephesos),3 as well as references to the roles of specialist interpreters (Dillon 2017, pp. 139–77). Beyond such texts, there are exceptional literary works, such as Aristophanes’ comedy Birds, which is full of references to observed species, but this is a play that is mainly interested in the comic possibilities of intelligent, human-like birds, or bird-like humans. There are also many images of birds, most of which are rather generic, so not easy to identify according to species.

By focusing on one region, the north Aegean area that was designated Macedonia and Thrace, in a defined period of time (6th to 4th centuries BCE), it becomes possible to make connections between birds that were consumed, and have been identified from archaeological excavations, and imaginary or imagined birds, and the ways in which the practice of divination gave meaning to two parallel kinds of activity, namely the observation of birds in flight, and a way of understanding the metaphysical universe. By studying these two separate but evidently related activities (observation and abstract thinking), we can give meaning to the kinds of entanglements that have united people and birds in historical contexts. Although archaeological evidence of birds being hunted and consumed is comparatively rare, the evidence from selected contexts at Pistiros, and other sites in Thrace, offers useful preliminary material for further study.

Bird bones consumed by ancient societies are rarely documented. From the excavations at Pistiros in central Bulgaria, there is a small but useful collection of bird bones, some of which have now been documented. The results show that the breeding of poultry, overwhelmingly chickens (68.9% of the specimens examined), but also geese and ducks (6.9%), was well developed from the early days of the city in the second half of the 5th century BCE (Boev and Stallibrass 2020, pp. 45–6). Most of the birds that people at Pistiros came across were domesticated fowl, which do not often appear in art, certainly north of the Aegean coast. There is a similar preponderance of domestic fowl over wild species at Olynthos (Stallibrass in Nevett et al. 2020, p. 352). Wild birds, particularly large birds, such as cranes and buzzards, and medium-sized birds, including grey rock partridges, were among hunted species at Pistiros, albeit representing less than a third of all bird bones. Their presence in the remains of human habitations is intriguing and has not so far been adequately explained. Cranes were among the birds particularly associated by ancient writers with Thrace (Aelian NA III.13; Mynott 2018, p. 10). Whereas bones of the white stork (Ciconia ciconia), which nests on rooftops, as well as in trees, might be expected amongst the residential structures at Pistiros, cranes prefer sheltered, waterside locations, reeds and rush beds. The presence of the ring ouzel (Turdus torquatus), a type of thrush, and more especially the common buzzard (Buteo buteo) and the hooded crow (Corvus cornix), are surprising in an urban context.

Zlatozar Boev, who has done more than any other scholar to identify and document bird bones in the south-east Balkan region, identified 78 bird taxa from archaeological sites in Bulgaria, including a handful of locations that belonged to the second half of the first millennium BCE and the first millennium CE. The pattern noted at Pistiros is consistent with the record found at other regional centres of population of this era. Domesticated fowls represent well over half the bird species recognised (including doves), but game birds (capercaille, pheasant, grouse) and waterfowl were present in appreciable numbers in many of the sites sampled. Birds of prey represent a distinctive sub-group in more than one third of sites studied. They included the common buzzard, tawny and little owl, several kinds of falcon, bustards (Otis tarda; Otis tetrax) and eagles, including the golden, Bonelli’s and short-toed snake eagle (Boev 1996, pp. 71–72; cf Boev 2006, pp. 109–20). The presence of raptors could indicate the beginnings of falconry, which is well documented from the Medieval period onwards, but may well have begun much earlier (Boev 1996, p. 76). The evidence of falconiform and strigiform raptors at many different settlements of the first millennium BCE shows that the presence of such birds was not random, but connected to some specific strategy, which might have included ornithoscopy but evidently did involve ways of keeping or securing these birds. Recent research on the cultivation of relations between human agents and raptors shows that various practices to secure birds of prey can be traced from the Neolithic period onwards, and perhaps even from Palaeolithic times. Falcons were caged in the cults of Re and Horus in Dynastic Egypt (Ikram 2023), while eagles played an active role in funerary cults and the social symbolism of first millennium BCE Kazakhstan (Lymer 2023). It would not therefore be surprising to find special relationships being developed with birds of prey in Thrace in the first millennium BCE. The epigram discussed below that describes the funerary memorial for a seer named Strymon provides exactly the sort of evidence that we might expect for a falconer.

Corvids, in addition to aquatic birds and birds that could be caught in shallow water, are particularly well represented among the large number of avian specimens from Bronze Age sites in the Aegean, a period that is presently better represented than the first millennium BCE on archaeological sites (Mylona 2022, App. A). The comparative data have furthermore confirmed that the number of domesticated species seems to have expanded noticeably in the first millennium BCE.

Scholars have only begun to understand the very long-term relationship between humans and birds comparatively recently, whether we consider wild or domesticated species. Without an extended period of acquaintance, in which humans tracked and trapped birds, whether to eat them, or simply to observe and try to domesticate them, there would have been no ‘ornithoscopy’. A recent reappraisal of the history of domesticated chickens in the Aegean area has shown how little we really know about these close interactions between people and birds. Chickens were first domesticated in south-east Asia, and the morphological changes associated with this process have been documented as far west as the upper valley of the River Euphrates in the second millennium BCE (Trantalidou 2013, p. 64 and n. 16; Clutton-Brock 2012 on domestication in general). Domesticated varieties appeared in the Levant and Anatolia much later, and in the Aegean and in Continental Europe at some point before the mid first millennium BCE. This is hard to specify, because traces of chicken bones have rarely been identified in archaeological excavations. Christophe Chandezon has presented such evidence as is currently available (mainly artistic and literary), which seems to show that although roosters were bred from the later eighth century BCE, domesticated chickens cannot be demonstrably identified before the mid 6th century onwards (Chandezon 2021, pp. 69–104, esp. 89, 95, citing Theognis of Megara [fl. ca. 540 BCE] who refers to cock crow, frgs. 861–64, Edmonds; see now Best et al. 2022).

Ancient preoccupations with fantastic creatures, composed of human and animal parts, require a different approach. Selected objects decorated with such creatures from the two case study regions offer further evidence of the kinds of human–bird ‘entanglements’ that can throw light on how people in antiquity thought about birds. One of the intriguing aspects of images and stories about birds in antiquity concerns the numbers of hybrid beings that appear to share bird-like characteristics, as well as anthropomorphic ones. What was it about birds that made them such attractive subjects, not just in their own right, as feathered creatures, but as models of supernatural abilities? This paper will explore some case studies, investigating how anthropological approaches to archaeological data can help to propose answers to such questions.

3. Orpheus and the Argonauts

One of the interesting observations that can be made when we select a narrower focus in time and space is that birds form a key theme in the association of ideas linking divination with narratives about Orpheus. These seem to form a discrete and inter-related set of ideas within the region and period in question. The earliest literary reference to Orpheus, by the poet Simonides of Ceos (ca. 556–468 BCE), also emphasises the singer’s evidently magnetic attraction to birds:

‘Numberless birds flew about over his head, and the fish leaped straight up from the darkling water at his lovely singing’.

(Simonides, PMG 567, tr. West (2005), p. 46)

Martin West used this fragment in his discussion of the possible existence of an early poem about the Argonauts, who included Orpheus in their number (West 2005). Several points in West’s paper are relevant for our consideration of Orpheus and birds. The first point concerns the fact that stories connected to the Argonautic legends underwent key changes before the first written version of the quest for the golden fleece emerged. West has argued that the earliest accounts of the Argonauts were oral only. In these orally transmitted versions, several elements were reported differently from those that became transmitted in the better–known, written accounts. In the oral version of the Argonautic legend, the ship’s passage through the Clashing Rocks, for instance, occurred on the return stretch of the journey, not, as later convention had it, on the outward leg. The later version evidently incorporated authentic, geographical knowledge of the problems of sailing into the Black Sea from the Aegean, and this is the version that we find echoed in the Homeric Odyssey, whereas the oral version(s) were more closely associated with the metaphysical dangers attending a hero’s return from ‘other worlds’ (West 2005, pp. 39–43). Homer’s Odyssey contains certain echoes of this orally transmitted version of the Argonautic story.4 West concluded that oral tales that were incorporated into the Homeric Odyssey reflect a combination of genuine echoes of pioneering journeys in the Black Sea, as well as into the western Mediterranean, combined with Indo-European folktale motifs and stories that are shared with the Babylonian Epic of Gilgamesh (West 2005, pp. 43–61, 62–64). I will return to the relevance of Indo-European folk tale elements below.

The prominence of extraordinary song, the characteristic that made Orpheus a key candidate for the crew of the Argo, is evident in one comparatively early piece of evidence, namely the metope showing the lyre-bearing Orpheus, who is clearly named, on a set of panels depicting the Argonauts on the Treasury of the people of Sicyon at Delphi (ca. 570–550 BCE: LIMC VII.I (1994) 81–105, #84 III.6: Figure 3). A new fragment from a papyrus roll penned by the poet Sappho also refers to Orpheus as a singer (Gronewald and Daniel 2005). Since Sappho’s birth is usually placed within a decade or so of the start of the 6th century BCE, this new fragment makes it potentially the earliest reference to the singer yet discovered. The lyric poet Ibycus of Rhegion (fl. 540/539 BCE) celebrated Orpheus as ‘famous Orphes’ (Fr.306, Campbell 1982), while the poet Pindar called Orpheus ‘the son of Oeagrus, Orpheus of the golden lyre’ (Pindar Fr. Thren. 3.11–12, Race 1997). These statements by poets from many parts of the Greek-speaking world from the mid-6th to the mid-5th century BCE show that the concept of the famed singer was widely established, even if he was thought of at this time (judging by Pindar’s evaluation) as the son of a Thracian nobleman, Oeagrus (cf. Pl. Symp. 179d; Apoll. Argon. 1.23–24; LIMC VII.I: 81–105; VII.2.57–77; Suppl. 1 (2009) 399–405). In Pythian Ode 4, Pindar included a few other details that localised some of the ship’s crew (Zetes and Calaïs, the two sons of the North Wind; Boreas, who came from the foothills of Mount Pangaeum: ll. 180–83), as well as stock on board (a red herd of Thracian bulls: ll. 205). Pindar’s story gives us the later version of the passage through the Clashing Rocks before the trip to Colchis for the golden fleece. Song, of a kind that could defeat most singers, remained the quintessential characteristic of Orpheus, well into the Roman Empire (e.g., Paus. 9.30.4: nightingales singing more beautifully on the grave of Orpheus than elsewhere).

If the earliest oral versions of the Argonautic story were circulating in the Aegean area before the Odyssey was written down, then these would belong in the 8th and early 7th centuries BCE, before the main thrust of Pontic travel and settlement by Aegean Greeks from the second half of the 7th century BCE, although current evidence suggests that this kind of traffic did not develop on any scale before the end of the 7th century (Monakhov and Kuznetsova 2017, pp. 65–66). It is therefore in the period after ca.600 BCE that we would expect the later versions of the passage through the Clashing Rocks to become established, but also better awareness to emerge among Aegean traders and travellers of the populations on the northern coastline, and in the direction of the Thermaic Gulf. This is also the period when lyric poets began to refer to Orpheus the singer and connect him with those very regions.

Much of the scholarly literature on Orpheus is concerned with his role as a teacher of rites and mysteries, a separate tradition that is apparent in parallel with his role as a singer. Herodotus was aware of this when he compiled his Histories: ‘It is contrary to [Egyptian] religious custom to be buried in a woollen garment or to wear wool in a temple. This agrees with the customs known as Orphic and Bacchic, which are, in reality, Egyptian and Pythagorean, for anyone initiated into these rites is similarly forbidden to be buried in wool. A hieros logos is told about these things,’ Hdt. 2.81 (tr. Graf and Johnston). Similarly, we find Euripides referring to the singer’s role in some of his plays (e.g., Bacch. 556), and to rites or mysteries in others (Alc. 328; Hipp. 936). The different traditions about Orpheus and his legacy have worried scholars, who have wanted to understand the evolution of these ideas in space and time (see further in Section 4, below), but given that the myths about Orpheus cannot be fixed in ‘real time’, the stories about the singer, like many myths, need to be understood as retellings or re-interpretations, recurring in different contexts and at different times depending on who was telling them and what social purpose they served.

The connection that I seek to make here is between Orpheus and birds on the one hand, and the divinatory practices of Thrace and Macedonia on the other. Herodotus refers to an oracular sanctuary of Dionysus in Mount Rhodope (7.110–112), situated on the highest mountain of the range, in the territory of the Satrae, but administered by the Bessi, one of the other tribal groupings from central Thrace, rather than from among the named coastal peoples. The historian adds that the oracular responses are delivered by a priestess, as at Delphi, and that the process is no more complicated at this sanctuary in Thrace than it was at Delphi (7.112). This seems to be a statement that responds to claims of something rather more mysterious. A slightly different clue is offered in a passage of Macrobius, who cites an Aristocles, in stating that an oracular sanctuary among the Ligyrians (?) in Thrace involved the process of drinking unmixed wine (Macrob. Sat. 1.18.1). The ‘Ligyrians’ in question are otherwise unknown, and the only plausible connection that has been made is with the River Lyginos, mentioned by Arrian, during Alexander the Great’s campaign towards the Danube in 335 BCE (1.2.1: Iliev 2013, pp. 65–66). Reports by ancient Greek and Latin writers about oracular shrines in the region of Thrace and Macedonia are few and the details in them are rather imprecise regarding the geography of the places and peoples referred to and about the kinds of activities that went on there. It is quite unclear how many oracular sanctuaries there may have been. A prophetic sanctuary required considerable investment, as well as the provision of personnel, to receive visitors and to furnish the services that clients expected. The reports that survive give modern readers no inkling at all about such measures, or about the functions of these oracular centres.

We thus know comparatively little about the divinatory practices of the communities of the east Balkan area, apart from a few textual references of this kind. One especially useful piece of evidence comes from a recently discovered papyrus manuscript containing epigrams very likely by the poet Posidippus of Pella, whose period of active production is dateable to the first half of the 3rd century BCE. One group of epigrams is entitled ‘Oionoskopika’ (Bird Divination). Dillon describes the fifteen epigrams as ‘the most extensive corpus of omens collected together in Greek literature’ (Dillon 2017, p. 142; cf Baumbach and Trampedach 2004).

One of the epigrams (no. 35) includes a commemorative vignette on behalf of the Thracian seer, Strymon (named after the River Strymon in Thrace), which includes the image of a raven on his memorial and refers to Strymon as the ‘Thracian hero, foremost overseer of bird omens;/Alexander dubbed him with his own seal,/for communing with his raven, he thrice defeated the Persians’ (Dillon 2017, pp. 143–44, tr. Dillon; Bastianini and Gallazzi 2001, pp. 148–50). Here we have the clearest indication so far of active engagement with bird divination in a context within our period, namely the campaigns of Alexander the Great (333–23 BCE), and their aftermath in literary commemoration. Posidippus came from this region and has left a vivid set of images of seers from the lifetime of Alexander that Great that show not just the importance of specialist ’bird lore’, but how the interpretations of seers were deployed in political contexts (Dillon 2017, pp. 142–55). There were routines, and there were consultations about specific moments in time, but those consulting, like Alexander, were not obliged to heed the words of their expert seers, and sometimes chose to ignore or side-step them.

Divination operated in different ways for different seekers of knowledge about the future. Posidippus’ epigrams reflect distinguished clients, and a few historical accounts echo similar requests at oracular sanctuaries within Macedonia and Thrace. In a celebrated passage, Suetonius described the visit of Gaius Octavius, father of Octavian, later Emperor Augustus, to a sanctuary of Dionysus (Liber Pater) during a campaign in Thrace in 59 BCE (Suet. Aug. 94.6). The location of the sanctuary is not given, but the prophecy accorded to Octavius was thought to refer to the future rule of his son. The fruitfulness or success of Octavian’s power was signalled by a huge flame at the sanctuary’s fire, larger than any that had been seen since Alexander the Great visited the shrine. Here, we learn about Alexander’s visit, as well as that of Octavian, and the report of a fire altar, connected to the form of prophecy practised there. The assumption seems to be that the quality or strength of the flame indicated the kind of answer to the question sought (Iliev 2013, pp. 65–66).

Two scholia to Euripides’ plays, one on Alcestis (schol. Ad Eur. Alc. 968), the other to Hecuba (schol. ad Eur. Hec. 1267), refer to oracular sanctuaries of Dionysus in Thrace. They both mention ‘writings of Orpheus’ kept at this sanctuary, referred to as a manteion, an oracular shrine (Iliev 2013, pp. 62–63, with discussion). Scholars have been sceptical of these references to ‘Orphic writing’, assuming that what was meant by the scholiast were extended writings, comparable to the type of text represented by the Derveni Papyrus, or at the very least, phrases and words that are found on gold lamellae in Macedonia (see Most 2022; Boufalis 2022). The lamellae are usually found in burials, which are understandably arranged to facilitate the passage of the deceased into the afterlife. Funerary practices may well have involved a set of ideas distinct from divinatory activities in life.

There may be a different explanation of the supposed Orphic ‘writings’, based in part on the materials referred to in Euripides’ play, Alcestis, which mentions sanidae, usually translated as wooden boards or writing tablets:

‘I have found no cure, whether in Thracian tablets, such as the voice of Orpheus wrote down’.5

Until the discovery of inscribed bone tablets at Olbia, southern Ukraine, bearing the abbreviated names of divinities, as well as the unexpected term ‘Orphikoi’ (followers of Orpheus), there were few obvious models for the reference in Euripides’ Alcestis. The Olbian bone tablets also include words such as ‘truth’, ‘life’, and ‘death’, as well as ‘Dionysus’ (Graf and Iles Johnston 2013, pp. 214–17). Knucklebones, used in divination by ordinary visitors to oracular sanctuaries, have been found in their tens of thousands in the Corycian Cave at Delphi, attesting to the enormous popularity of this method of seeking prophetic information (Amandry 1984). Of the 22,771 mainly sheep and goat knuckle bones identified by researchers, the vast majority were unmodified (18,709); the remainder were worked, pierced, gilded, or had lead added to them, or were inscribed. Altogether, 31 were inscribed, usually with abbreviated names of gods or semi-divine beings, including Heracles, Thetis, Achilles, and Ajax. The graffiti on the Olbian bone plaques do resemble these kinds of inscribed names at Delphi, and at other sanctuaries. It is possible that the reference in Euripides’ play Alcestis is to bone plaques, or tablets, like the Olbian ones, rather than to conventional writing tablets. The words that appear on knuckle bones (astragali), often in abbreviated form, represent a close conceptual link with the Olbian plaques (Domaradzka 2013, pp. 205–6).6

4. Classifying Bird Hybrids in Antiquity

Birds have particularly attracted the attention of classical scholars. Among the earliest modern studies of animals in the remote past was D’Arcy Wentworth Thompson’s A Glossary of Greek Birds (Oxford, 1st ed. 1895). The author, a professor of natural history at Dundee, laboured to identify species described by Aristophanes, Aristotle, and other Greek and Latin authors. The dozens of ornithological species referred to by Aristophanes in his play Birds is one reason for this scholarly enthusiasm. Aristophanes’ imaginary creator of a ‘bird world’ in the heavens, a man called Pisetaerus, is aware that everyone around him seems to want wings, and jokes that words can ‘help everyone take wing’ (Arist. Av. 1438–1439). Aristophanes’ play Birds was produced in Athens in 414 BCE. References to winged creatures, or metaphors involving wings and flight, are also quite common in 5th century BCE tragedy.7

The reader of Aristophanes’ Birds gets a vivid idea of how much thought went into the poet’s metaphors and allegories about a bird world modelled rather closely on the world of humans. Two mythical characters who are given considerable prominence in the play, Tereus and his wife Procne, were otherwise represented in an earlier tragedy of horrific intensity (so far as the fragments allow us to judge), created by Sophocles, in which they were transformed, respectively, into a hoopoe and a nightingale. Aristophanes parodies his older contemporary by making Tereus a thoroughly Athenian-seeming and sounding bird. He was formerly a man, and now becomes the intermediary who can help in communications between the bird world and humans. In the legend reworked by Sophocles, Tereus was punished for his crimes against his wife Procne and sister-in-law Philomela by being transformed into a hoopoe. Coincidentally, Sophocles made Tereus a Thracian king. Aristophanes took Sophocles’ material and made something completely different with the story about metamorphoses into birds. At the same time, the comic playwright appears to be rejecting the negative associations that Sophocles seems to have implied between barbarian Thracians and cultivated Athenians, even if the outcome of Sophocles’ drama was more complex than such a straightforward polarity suggests.8

How should we interpret these rather varied and quite complex ideas about birds? Jeremy Mynott, in a recent book that surveys birds in Greek and Roman culture, thought that the ubiquity of birds, and bird-like beings, can be explained by the sheer numbers of birds in the remote past, and by the ways in which birds had an active presence close to human societies (Mynott 2018, pp. 357–58). Modern scholars are aware that birds, like insects of many kinds, were much more numerous in past centuries, and the ubiquity of birds may indeed have been a factor in ancient authors’ preoccupations with them. Nevertheless, numbers alone cannot account for the specific symbolic roles given to birds in the ancient societies of Eurasia. Birds appear in the artistic repertoire of many ancient societies, which suggests that we also need to think more broadly about the mutual interactions of humans and birds in the remote past, rather than solely through the lens of given ‘cultures’, as the data are usually presented to readers (See e.g., Walker 1996; Raja Dinkar Kelkar Museum 1988; Lev-Tov 2004; Gunter 2019).

The idea that creatures who were conceived of, in some sense, as half-way between the divine and the human (whether physically, metaphysically, or metaphorically) should be bird-like, and be able to fly, is already embedded in the Homeric poems of the 8th and 7th centuries BCE, and remained a vivid literary and artistic trope thereafter (Petridou 2015, pp. 87–98; Dillon 2017, pp. 139–77; Mynott 2018, pp. 246–66, 304–334; Bremmer 2020, pp. 104–25; Kindt 2020b, pp. 197–216). Mynott accepted the view of the anthropologist Claude Lévi-Strauss that birds, like other animal species, are ‘good to think with’ (‘Les espèces sont choisies non comme bonnes à manger, mais comme bonnes à penser: Lévi-Strauss 1962, p. 128, English tr. 162; Mynott 2018, pp. 243, 245). Lévi–Strauss’s perceptive formulation has been applied widely in analyses of ancient cultures and deserves to be underscored in the present context. It is a formulation that helps us understand how it was possible, on the one hand, for Aristophanes to name, and include among those appearing on stage in the Birds a range of evidently well-known birds, as well as exotic birds, who mimic human behaviour and decisions in a delightful, satirical mode, whilst also accepting the possibility that all kinds of extraordinary bird-like creatures could exist. Among the identifiable species of birds that the poet mentions are the crow, jackdaw, hoopoe, nightingale, wren, peacock, marsh bird, black partridge, vulture, bunting, gallinule, wild pigeon, kite, swallow, cuckoo, eagle, coot, duck, seagull, stork, swan, hawk, ostrich, pelican, marsh harrier, tern, Marsh tit, Great tit, thrush, blackbird, owl, corncrake (?), shelduck and quail. The ‘Persian bird’, or chicken, is also mentioned (Ar. Av. 485, 707, 833–836), and is a very rare reference in 5th century BCE literature to this domesticated species of bird. At the same time, there are in the play creatures, whether birds or half-bird, half ‘imaginary’ mammals, that also find their place, such as the halcyon (Ar. Av. 251, 1594, usually identified by modern scholars with the kingfisher) and the ‘hippalektryon’, or horse-rooster (Ar. Av. 800; Dunbar 1995, ad 268–302; 249).

Semi-anthropomorphic winged beings are almost taken for granted in the scholarly literature on ancient mythology, whether it be Eros, Victory, Sleep or Death, perhaps in part because of the sheer volume of images that survive, particularly those of Eros and Victory.9 The category of ‘half-bird’, ‘half–human’ being is, in fact, an awkward one to interpret in a setting of human–divine relationships. Few winged divinities received cult recognition (Aston 2011, pp. 133–36). In trying to locate these winged beings in a cult environment, Emma Aston reflects that wings facilitate flight, and aid pursuit (in the case of divine amours), but also that ‘bird identity is in the mixture too’, citing the Sirens as an example of an imaginary type that cannot be comfortably classified (Aston 2011, p. 136).

The range of nuances that the concepts that half-human, half-bird images were intended to convey evidently needs to be enlarged. Most recent papers and books have focused on references to birds in ancient literary texts that can be identified using modern classifications (see esp. Pollard 1977; Arnott 2007; Kitchell 2014, passim; Dillon 2017, pp. 143–58; Mynott 2018). The merit of such textual surveys is that they offer a vivid and quite precise idea of observed birds, and some initial ideas of ancient intellectual responses to many kinds of birds, particularly those observed in urban settings, on agricultural land, and in lacustrine environments, as the list above, gleaned from Aristophanes’ play Birds, reveals. The disadvantage, however, is that the focus on literary references obscures forms of behaviour that are not reflected in literature, including many of the more mundane forms of divination, to say nothing of the ordinary practices in and around bird domestication and bird hunting. Understanding half-bird, half-human images and ideas takes us beyond the scope of such straightforward classificatory investigations.

The kinds of ideas that Aristophanes was presenting to his audience here offer us a many-layered, sophisticated example of human–animal ‘entanglements’. ‘Entanglements’ is a term originally applied to quantum physics which has been adopted by anthropologists and social historians and has been transferred by Jeremy McInerney to the world of human–animal relations in antiquity (McInerney 2020). In physics, the word ‘entanglement’ was intended to convey the behaviour of sub-atomic particles, whose properties are hard to define except by reference to other particles. McInerney wants scholars to consider that human relations with animals should be similarly conceptualized. Animals are among the oldest artistic representations by anatomically pre-modern as well as modern humans, and these include the extraordinary ‘Löwenmensch’, the ‘lion-man’ figurine in mammoth ivory from the Stadel Cave, Hohlenstein, Baden-Württemberg, Germany (35,000–40,000 years BP). This small (31cm long) sculpted figure has a lion’s head but stands upright on two legs, which resemble the legs of a man, but have the powerful and sleek musculature of a big cat. Human and animal seem to be strangely combined. McInerney refers to the ‘Löwenmensch’ so as to reinforce his thesis that human societies cannot be studied apart from animals, or the other way around. Aristophanes’ comic ideas seem to confirm this way of thinking. The play Birds may be a particularly telling example, which was sufficiently enthralling to have found its way into artistic representations (Green 1985). At the same time, the ‘Löwenmensch’ points to a different way of understanding the half-bird, half-human images that have not been fully explained in earlier studies of Eurasian cults or funerary practices.

In The Origins of Monsters: Image and Cognition in the First Age of Mechanical Reproduction, David Wengrow explains the persistence and reproduction of composite animals as the result of subtle cognitive processes in the human imagination, or ‘complex conjunctures of social, technological and moral processes’ (Wengrow 2013, p. 112). The stories about non-human animals that circulated in ancient societies had deep origins, and the ideas about terrifying, retributive beasts did not go away with the emergence of ‘rational’ philosophies. So much is clear from the contents of the papyrus fragments, containing the earliest surviving evidence of philosophical speculation of an ‘Orphic’ kind, found in tomb A at Derveni, Macedonia. In the deciphered text, ‘Erinyes’ (Furies) and ‘daimones’ (spirits) appear repeatedly in different sections of the two latest transcriptions, by Richard Janko and Valeria Piano (Most 2022). Despite the co-occurrence of these different types of creatures, or zoomorphic hybrids, scholars seem to persist in seeing the observed creatures as a reference to ‘rational’ discourse, and the hybrid forms as ‘irrational’. In one of the most recent compilations of animal studies, Julia Kindt’s (2020a) edited volume Animals in Ancient Greek Religion the renowned scholar of ancient religions, Jan Bremmer, says, in a paper on theriomorphism: ‘One cannot escape the impression that the divine theriomorphic shape and moment belong to an older layer of Greek religion, even as that layer survives in numerous later texts and areas’ (Bremmer 2020, p. 112). Whether or not animal hybrids should be seen, in some sense, as the survivors of some ‘older’ perception of the divine, their potency was still apparent, as the representations in the objects illustrated in this paper show.

Perhaps there is some truth in this, insofar as certain Indo-European mythological themes seem to have survived, as we shall see below, in an abbreviated or reduced form, by the 5th and 4th centuries BCE, which is the time frame within which reflective ideas about myths were first written down in a form accessible to us. However, it would be a mistake to think that demons or monsters had been banished to the fringes of contemporary culture. We will need to reconsider David Wengrow’s observations about the ‘conjunctures between social, technological and moral processes’.

Yet Wengrow’s is not the kind of hermeneutic approach that has been applied to 1st millennium BCE case studies. The relationship between humans and animals in antiquity is a topic of research that has undergone a veritable transformation in the past decade, with the appearance of major survey works (those in the English language include Campbell 2014 and Lewis and Llewellyn-Jones 2020) as well as important theoretical reviews (Kindt 2017). Birds are prominent in these works, and in dedicated studies of Greek divination, as witnesses or agents of divine communication with humans (Burkert 2005, pp. 29–35; Dillon 2017, pp. 139–77; Bremmer 2020, pp. 104–14; Kindt 2020b). Such studies of divination usually understand the process in terms of the observation of real birds in flight, from specific viewpoints in a given spatial setting (an oionoskopeion: Dillon 2017, p. 5, 46 n. 22, 153–4; Kindt 2020b, pp. 200–2). The rationale behind this mechanical process is that the ‘natural order’ provides ways of understanding and articulating supernatural knowledge (Kindt 2020b, p. 200). However, as Burkert observed, the relationship between observation and interpretation is a complicated one, and human beings who try to test divine providence by pressing for specific answers seem to get into difficulties (Burkert 2005, pp. 36–49).

The class of half-human, half-animal being appears as something of an afterthought in the Lexicon Iconographicum Mythologiae Classicae (LIMC 1981). This class certainly deserves greater scrutiny. Animal hybrids are listed under ‘Monsters’, a rag-bag of a term that has been used to describe creatures that cannot otherwise be categorised (LIMC I (2009) 339–71, sv. Monstra; cf Daemones anonymi (Etruria), pp. 143–56). These ‘monstrous’ beings include unusual human–animal hybrids, such as centaurs, sphinxes, sirens, the Chimaera and Hippalektryon. It is this very category that Emma Aston chose to put into the limelight in her 2011 book, Mixanthrôpoi. The origins of many if not most of these ‘demons’ lies (as far as we can observe from surviving imagery) in Mesopotamia, in the Neo-Babylonian and Neo-Assyrian artwork of the first half of the first millennium BCE, which was disseminated probably through seal- and gem-stones (such as the one illustrated in Figure 4a,b), but also textiles and other more ephemeral artefacts (Black 1992; Counts and Arnold 2010). In Figure 4a, we see a ram-headed sphinx, a distinctive combination of animal forms that is found in Egypt but nowhere else in Eurasia. The reverse of the same seal-stone shows a falcon-headed sphinx, another specifically Egyptian variant (Figure 4b). In the western Mediterranean, winged demons with anthropomorphic bodies are very apparent in Etruscan art, and it may be the case that there are cultural reasons why similar demons occupy a different place in the artistic production of mainland Greece.10

5. Winged Creatures, Severed Heads in Thrace, and Orpheus Again

Among the items listed in the Lexicon Iconographicum under ‘Monsters’ is a silver jug from the hoard of silver vessels from Rogozen, in north-west Bulgaria (LIMC Suppl. 342 no. 7; Figure 5). The body of the vessel is decorated with repoussé figures. At the front is a ‘potnia thērōn’, a winged female divinity holding two dogs by the fore paws (Figure 6). Either side of her there are winged centaurs with human heads. There is a mythological tale behind these figures that is now unknown to us. Nevertheless, the ‘potnia thērōn’ is a figure that was widespread in the ancient world, with a strong pattern of artistic representations in the first millennium BCE, extending from Mesopotamia to central Europe (Counts and Arnold 2010, pp. 11–20). These images reinforce the arguments put forward by linguists and comparative cultural historians about the distribution of Indo-European folklore elements that underlie the narratives echoed in Greek literature of the same period (West 2007, pp. 20–5, 152–53, 191–93, 468–70). Another jug from the same Rogozen hoard shows a Chimaera, composed of a winged lion’s body, a griffin’s head, snake’s tail, and a female head behind that of the griffin (Figure 7). There are aspects of the Chimaera’s body that are familiar from Greek texts and images, but the woman’s head is something new. The presence of the goddess figure with animals and centaurs deep in the Thracian interior (Rogozen is in the far north-west of Bulgaria, in the Danube Foreland) shows that the images of the Mistress of Animals, like the centaurs, and other figures of animals and birds represented in this rare collection of silver plate, combine ideas that we recognise in objects from widely separated regions. These figure-decorated objects from Thrace fill a geographical and cultural gap in the visual records from Eurasia. They show that there were stories in circulation containing Indo-European or wider Eurasian elements that link hunting with mastery over nature in a cosmological projection, not just as a functional activity (see esp. Arnold 2010, p. 195; Marazov 2011, pp. 138–46). In Neo-Assyrian sculpture, the imagined execution of a mythological beast by a hero or ruler expressed the correct world order of living things with nature. Hybrid creatures played a key part in these deeply encoded images of a cosmologically driven mastery (Ataç 2006, pp. 91–93). The precise significance for Thracian societies of the Rogozen material is less easy to extract. These pieces of silver show the thematic range of the stories that were circulating, and the discrete nuances that distinguish the artistic repertoire beyond the northern Aegean coast from the Greek-speaking Aegean (Martinez et al. 2015, p. 342 [K. Rabadjiev]).

If there is one distinguishing feature of these images, it is the presence of animals and birds, creatures that are directly linked to Orpheus in texts (e.g., Paus. 9.30.4) but rarely represented in Greek images of the singer. The Rogozen vessels, on the other hand, are simply teeming with zoomorphic figures, and with human–animal hybrids. Orpheus may not be present, but his creatures are. In these images, the Near Eastern origin, or inspiration, of the hybrid forms is apparent in the number of wild species represented, in the preoccupation with hunting, and in the combinations of anatomical elements from different wild creatures, including a griffin with the body of a winged feline, goats or ibexes, and the strange raptor bird with a snake in its mouth (cf. Marazov 2011, pp. 147–81; Spier et al. 2022, pp. 95–114, nos. 12–28, 35–43, Achaemenid silverware, jewellery and sealstones). The clear focus on wild species is characteristic of the Continental European, Eurasian and Near Eastern repertoires of artistic images but is absent in mainland Greece (Arnold 2010; O’Sullivan and Hommel 2020, pp. 54–66). The question of how Orpheus was fitted into the zoomorphic menagerie is too complex to be dealt with here (Jesnick 1997 for images); I will focus on the winged hybrids.

The animals and birds are part of a visual language with a wide currency in the Continent of Europe, and in Eurasia, in which divine beings could take animal and bird forms (Figure 8), a pattern that is less often found in Greek tradition (Bremmer 2020, pp. 102–25). Rogozen jugs no. 158 (Figure 5 and Figure 6) and no. 162 (Figure 7) show an unusual number of winged beings. On no. 158, the Mistress of Animals is winged, as are the centaurs. On no. 162, the Chimaera has wings. These additions to the commoner, wingless versions of these creatures give them special, semi-divine status, which seems particularly unexpected in the case of the Chimaera. The wings confer a more terrifying significance to these hybrid beings. They remind us that winged creatures can denote negative as well as positive power.

In Egypt, the Ba-bird, which represents the non-physical properties of the deceased, was very likely the direct model for the siren, which, in the wider Mediterranean, is regularly connected to cult dedications, as well as to death and mortuary practices (Tsiafakis 2001, pp. 7–13).11 Egyptian Ba-birds also accompany offerings to the gods (Figure 9).

In the wider Mediterranean, sirens were separate beings from deceased individuals. Sometimes they seem to be benign, and at other times menacing. The creatures on the Rogozen silverware have a similarly ambiguous quality. We are reminded of the ambiguous nature of the ‘demons’ and ‘Erinyes’ in the Derveni papyrus texts. Although Pausanias tells us almost nothing of value in the Guide to Greece about the birds and animals that found Orpheus’ magnificent songs irresistible (9.30.4), he does make the extraordinary remark that Orpheus derived his exceptional power from a common belief that he had discovered mysteries, such as how to be pure from sins, how to cure diseases, and (most particularly) how to avoid divine wrath. The Derveni texts are very much preoccupied with this last fear and some, at least, of the artefacts that we are looking at may well have had a similar purpose.

Among the Rogozen silver vessels is a jug (no. 159) which shows a boar between two mounted hunters. Beside one of the hunters is the severed head of a woman. Nancy Thomson de Grummond has identified this as a prophetic head. She connected this head with the Orphic and Pythagorean idea that the soul enters the human body through breath (Thomson de Grummond 2011, pp. 332–33 and Figure 10.12; 342 n. 63). There are severed heads on Etruscan mirrors which seem to allude to the iconography of Orpheus (particularly those heads shown wearing a Phrygian cap), and some are winged heads (Thomson de Grummond 2011, pp. 320–31). She compares the Etruscan and Rogozen images of severed heads with a series of gilt silver plaques from Letnitsa, north-west Bulgaria, where severed heads of a man, a woman, and a horse appear multiple times. The severed heads are slightly larger than the heads of other figures depicted on these plaques, and the mouth of the severed head appears open wide, as if singing or shouting.

In his introduction to a paper on the incidence of the severed head in the mythology of Orpheus, by Joseph Nagy, Lowell Edmunds points out that recognisable Indo-European story motifs are far less prominent in Greek myth than one might expect; in some sense, they seem to be backgrounded and can be identified in stories that were perhaps less popular or in versions of those stories that were less commonly repeated in Greek-speaking contexts (Edmunds 1990, pp. 199–200). Nagy has studied the stories about Hercules, and particularly about the head of Orpheus and its oracular properties after death. Many of the stories about the head of Orpheus are reported in comparatively late Greek literary accounts (e.g., Conon FGrH 26 Narr. F45: Brown 2002, pp. 302–8), and major writers, such as Pausanias, fail to mention the severed head at all (9.30.3).12 There are various tales in Continental Europe about the later history of the/a severed head belonging to a significant bard, usually intended to be perceived as a foil to an armed hero. The rare references in Greek literature, and occasionally in ancient Greek art works (LIMC VII.I (1994) 88 # V.68–70; Dillon 2017: 63 Figure 2.3 [Cambridge]; 64 Figure 2.4 [Basel]; 117 Figure 3.4), to a singing, severed head, can be paralleled in Norse, Irish, and Indian folklore traditions (Nagy 1990, pp. 208–28). The location of Orpheus’ death is nevertheless always placed either in Leibethra, Macedonia, or in Thrace. Sometimes there is confusion (as, for example, in Apollonius’ Argonautica (1.23–24)), where place names in Macedonia and in Thrace have become mixed up.13 To some extent, therefore, the public conversations about Orpheus and his cultural origins were already opaque by the 3rd century BCE, and most of the written accounts about him postdate that era. Such associations as had been made by authors of the 6th and 5th centuries had already become erased and separated from the kinds of stories heard, or observations made by contemporaries, based on experiences (of whatever kind) in the northern Aegean. After the 4th century BCE, the connection between Orpheus and Thrace, or Macedonia, seems to have become irrelevant to writers.

The local stories about Orpheus and his prophetic qualities must be rediscovered in other ways. In Thrace, we find the severed head of a man, perhaps Orpheus, on gilded silver plaques from Letnitsa, in Bulgarian north-west Thrace, as previously mentioned (Martinez et al. 2015, Cat. Nos. 293, 294, 296 ‘mid 4th century BCE’), and on harness attachments from Vratsa, in the far north-west of Bulgaria (Martinez et al. 2015, p. 278, No. 239, ‘third quarter of the 4th century BCE’). The burial of a bronze portrait head that had been severed from a full-size statue, in the entrance to a tomb, and of a historical ruler at that, Seuthes III, has clear echoes in the Medieval Irish and Norse sources discussed by Nagy (Martinez et al. 2015, p. 119, Cat. No. 82, ‘final decade of the 4th century BCE’). The practices enacted at the tomb of Seuthes III also reflect Conon’s account of the death of Orpheus, in which the head was buried, following its arrival at the mouth of the River Meles, below a mound, and became a hero-shrine thereafter (Conon FGrH 26 F45; Nagy 1990, pp. 211–20). In Conon’s tale, the burial of the head is linked with the expiation of guilt by Thracians and Macedonians for the murder of Orpheus.

These examples of severed heads, projected into a narrative and ritual context, have survived in northern parts of Thrace. The written accounts of Orpheus’ prophetic, singing head, which are situated in myth in the estuary of the River Hebros, or in the north-east Aegean,14 offer a southern counterpart of the images on gilt silver vessels and harness equipment from north Thracian burials. The myths that associated a severed head with oracular properties, and are reflected in art works, particularly in Thrace, but also in Etruria, show that these stories were embedded in Continental Europe, rather than in the Mediterranean (Thomson de Grummond 2011, pp. 315–35). Nancy de Grummond’s exploration of Etruscan images expands and augments the literary accounts studied by Nagy. These Continental traditions about oracular severed heads show the penetration of Eurasian concepts of prophecy, which in Thrace are linked to the observation of and sequestering of raptors, in the direction of the Mediterranean. A parallel cultural trend saw Near Eastern conceptualisations of demons moving in the opposite direction, from Mesopotamia via Egypt, into the Mediterranean. We see images of winged creatures, some of which were entirely imaginary hybrid beings, but could have human parts. This trend explains the appearance of sirens, as shown in Figure 1 and Figure 2, but also the strange composites shown in the Rogozen jugs (Figure 5, Figure 6 and Figure 7).

The historian Herodotus (2.81) consciously associated rites performed by Orphic followers with Pythagoreans (in southern Italy) and with an Egyptian source. In other words, in the second half of the fifth century BCE there was an awareness, among speculative thinkers at least, of the extensive geographical scope of ideas that connected the ethical behaviour of the living with their outcomes after death. The dispersal of seers, winged hybrid beings, and of severed heads in artworks across the Mediterranean shows how widespread at least some of these ideas were, however superficial may have been their ‘Orphic’ connection. Nevertheless, the observation of birds is implicated in these ideas, in a variety of ways, and (according to Greek writers) underlies the motives behind ornithomancy.

6. Conclusions

In antiquity, we find a wide range of cultural applications of birds and bird-like beings. The most significant applications seem to have concerned the observation of birds in divination rituals. Alongside these divinatory practices, we find bird-like images of composite beings, whose existence seems to iterate and expand upon the range of human cognition of birds. This range of cognitive appreciation can be studied by combining evidence from ancient literary and epigraphic texts, archaeological and zooarchaeological data. By selecting one geographical region (Thrace and Macedonia) in one specific time period (the 6th to 4th centuries BCE), it has been possible to isolate particular instances of bird–human ‘entanglements’ which help to contextualise these relations. By focusing the cultural range of investigation on a limited area, it has been possible to make connections between artefacts and ideas and give added resonance to images involving birds or bird-like creatures.