And trusting and strong in faith he had gone about his business, the while a black shadow hung over him and a horrid Fate waited in his pathway. Now suddenly it had swooped upon him, and had seized him by the leg. Relentless, remorseless, it was; all his protests, his screams, were nothing to it – it did its cruel will with him, as if his wishes, his feelings, had simply no existence at all; it cut his throat and watched him gasp out his life. And now was one to believe that there was nowhere a god of hogs, to whom this hog personality was precious, to whom these hog squeals and agonies had a meaning? Who would take this hog into his arms and comfort him, reward him for his work well done, and show him the meaning of his sacrifice? Perhaps some glimpse of all this was in the thoughts of our humble-minded Jurgis, as he turned to go on with the rest of the party, and muttered: “Dieve – but I’m glad I’m not a hog!”
Upton Sinclair, The Jungle

1The first spell from the first of the Greek Magical Papyri contains a mystery: “take a circaean falcon,¹ and deify it in the milk of a black cow, mixed with Attic honey, and when it is deified wrap it in an uncoloured strip of cloth… and set it up in a shrine of juniper”.² A footnote to the translation, by Robert K. Ritner, tells us that “[t]he magical rite of drowning effects deification”.³ This is a mystery in the ancient sense of the word: a secret and deeply significant rite which might nonetheless appear paradoxical or shocking to the uninitiated. This mystery – a drowned falcon worshipped as a god by his killer – is not alone in the corpus of Greek and Demotic magical papyri, but it is the first strand of a red thread that I will attempt to unravel here, a thread spun from Egyptian ideas about the lives and deaths of humans, animals, and gods.

2I will begin this discussion with an overview of magical texts in which animals are drowned, examining their treatment and role in these rituals, before looking at the role of animals in Egyptian religion more broadly, in particular the creation of votive mummies, and then of the religious significance of drowning in Graeco-Roman Egypt. If these two discussions put their focus on the second-order category of ‘religion’ rather than ‘magic’, they are nonetheless important for our understanding of the magical papyri, whose ritual mechanisms depend heavily on the theological concepts of the Ptolemaic and Roman temple cults. In the final part of this discussion I will turn to consider what these rituals tell us about the way in which the identities and personhood of human and non-human animals were conceptualised, discussing the cases of the Apis bull, the divinised Antinous, and a drowned cat.

Circaean falcons and cat-faced spirits

3The ritual involving the circaean falcon is only one of nearly thirty such recipes which survive from the Roman period, in which various animals are drowned in magical rituals. Twenty-one of these are found in seven papyrus manuscripts which survive from 3rd and 4th century Egypt,⁴ and a further eight occur in the Cyranides, a collection composed and collated in the first few centuries of our era, perhaps in Alexandria, and preserved in manuscripts dating from the 12th to 16th centuries.⁵ While falcons feature prominently in these recipes, there are many other animals – beetles, shrews,⁶ lizards, various kinds of fish, as well as a cat, a lizard, a swallow, a mole, and a hoopoe, appearing in one recipe each (see table 1). The purpose of this section will be to describe these rituals – the way in which the animals are killed, what is done to their bodies after their deaths, and the role they play in the rituals of which they are part.⁷

4As is common in magical texts, the length of the ritual prescriptions varies considerably; some are laconic instructions to simply drown the animals, others detailed, step-by-step procedures specifying what should and should not be done at each stage. Nonetheless, there are enough shared details to reconstruct an “ideal” ritual process, from which significant deviations can be noted.

5A few of the texts mention that the ritualist is to have undergone a period of purification prior to carrying out the ritual, as is common in Graeco-Egyptian magic; this consisted of avoiding sexual intercourse, meat, and bathhouses – impure places – for a fixed number of days.⁸ The next step, acquiring the animals, is elided in most of the rituals, but whether they were caught wild, bought from merchants, or acquired by other means,⁹ the next stage was to drown them. A wide range of liquids were used for this – water (often specified as spring, rain or river water), oil (which might be scented with roses or lilies), wine, honey, tree-resin, bull-fat – or some combination of these. There is to some extent a practical aspect here; liquids other than water were used to kill water-breathing fish, but the choice of liquid seems to have also played a vital role in the ritual mechanics: wine, beer, honey, and milk were the same liquids used as libations in traditional Greek and Egyptian cults, and in other magical rituals from this period.¹⁰ Unmixed wine, or mixtures of wine, water, milk, honey, and oil were traditionally used in Greek cults which honoured and nourished chthonic deities and the dead,¹¹ while these same liquids were associated in Egyptian culture with fertility and the Nile Flood, and were offered in temple cults.¹² Milk, in particular, was associated in Egyptian cults with child deities – it is, fundamentally, a substance which nourishes the newborn – and by extension the nourishment of the reborn dead.¹³ The “black cow” whose milk is called for in several recipes¹⁴ probably stands in for the mother goddesses who take this form: the wet-nurses of the souls of Heliopolis in the Pyramid Texts,¹⁵ and the primeval goddess Amaunet in a Roman-period cosmogony.¹⁶ The oil of lilies may draw upon similar associations, since the Byzantine-era Geoponica tells us that the lily sprang up from the spilt breast-milk of the goddess Hera,¹⁷ while the rose, which scented other oils, may have been associated with the crown of roses worn by the justified dead in Egypt.¹⁸ The liquids used were thus highly signified, linked to religious rituals in general, and in particular the sustenance of gods and of the newly (re)born.

6In several of the texts a spell is spoken over the animals as they were drowned,¹⁹ and the animals might be left in the asphyxiant – the liquid in which they had been killed – for a significant time after their death: two days in the case of a Nile Perch,²⁰ seven days in the case of a scarab beetle.²¹ In each of these cases the asphyxiant is oil, a liquid which would have prevented the corpse’s immediate decomposition. Somewhat strangely, one of the texts suggests that the animal – a beetle in this case – is dead before it is placed in the water; it is to be fed a cake made from lotus seed and honey, which is said to kill it.²² Perhaps in an earlier version of the text the cake was simply intended to drug it, to make its drowning easier, a possibility which – though unstated – we might well imagine in the case of larger animals such as cats and falcons, equipped with claws with which to resist their would-be killers.

7After being killed, the animal might be dried; two texts describe suspending a fish over the bowl in which it has been drowned for two days,²³ while another describes a scarab being dried on sand for four.²⁴ In other texts, this placing of the body on sand is delayed until after it is wrapped in the manner of a mummy,²⁵ either with the traditional linen,²⁶ or with a strip of papyrus.²⁷ This often consists of an elaborate procedure; the bindings may be decorated with magical formulae and drawings,²⁸ or else accompanied by strips of particular colours²⁹ – blue, red, white, and black, or white, green, blue and red – and the process might also involve the use of natron salt, with its desiccant and purificatory qualities, and the fragrant preservatives myrrh and frankincense, as well as saffron and wine;³⁰ these substances, sticky and perfumed, would presumably help to hold the bindings together, and mask any smell of rotting.

8It is true that this process bears little immediate resemblance to the best-known and most detailed accounts of mummification, provided by Herodotus and Diodorus Siculus,³¹ in which human subjects had their internal organs removed with varying degrees of care, and either stored in jars, discarded, or returned, wrapped, into the body cavity along with aromatics such as myrrh, and in which they were dried for seventy days with natron, before being bound with fine linen, and plastered with gum. But this is to confuse technical details with ritual efficacy; the well-known process described by Herodotus was apparently used only on the wealthy, and significantly less elaborate procedures could be used on humans and, significantly, animals.³² These might consist, as in the case of the magical mummies, of plastering them with fragrant resins before wrapping, or even simply providing them with basic funeral rites and amulets. In their specifics, the magical mummies seem to broadly follow those laid out by the most detailed surviving account of mummification, the Ptolemaic Apis Embalming Ritual,³³ the instructions for mummifying the divine Apis bull who resided at Memphis.³⁴ While the bull did undergo the removal of organs and drying process associated with high-quality human burials – the care taken in embalming this bull was (at least in theory) equal to that accorded to royalty³⁵ – it was placed on a bank of sand at various key stages of the embalming procedure,³⁶ and its wrapping involved the use of coloured cloth – blue and red, in this case.³⁷ The sandbank here, at 6 by 4 cubits in size, would have been far larger than that used for the magical mummies, but its ritual purpose was the same; it prevented contact with the earth, and, as a representation of the primordial mound, the first land formed by the creator in Egyptian cosmogonies, cast the process of mummification as a new creation, the resurrection of the dead into a new, divine existence.

9But while funeral rites of the Apis bull and similar animal gods were paid for by the king and temples³⁸ and represented a pious act on behalf of the kingdom, the creation of an animal mummy in the magical rituals was a private act, and this private aspect is reflected in the incorporation of the ritualist’s own materia into the process: parts of the ritualist’s body are used in the wrappings – two fingernails and all of their hair in one ritual, an unspecified number of nail parings in another, and blood from their thighs in another;³⁹ these same materia are often used in other rituals to create links between individuals, as intimate body-parts which can nonetheless be removed and manipulated.⁴⁰

10The Apis bull, and other divine and sacred animals, would be installed in tomb chapels or catacombs after their mummification, and this too seems to have been the case in several of the magical rituals: the vaguer instructions refer to the mummy being placed in a “hidden place” or near the dwelling place of the ritualist,⁴¹ while the most explicit, the ritual involving the circaean falcon with which we began, refers to the mummy being placed in a shrine of juniper, which is then crowned.⁴² The process seems to have ended, in the most elaborate rituals, with acts reminiscent of those carried out for the deified dead, human and non-human, more generally in Egyptian culture: a ritual meal, the burning of incense, and offerings of food and drink.⁴³

11This whole process clearly draws upon general Egyptian ideas about the afterlife; the concept that the individual survived after death, and that, with the help of funerary rites which made the body into an undecaying and ritually protected resting place, the individual could enjoy a divine existence, traversing different spheres as a transfigured spirit, returning to its body to rest and interact with those left on earth. This model, which had existed for thousands of years by the time the magical mummies would have been made in Roman Egypt, was based on two key archetypes:⁴⁴ the first of these was the sun-god, who nightly died at sunset, traversed the underworld, and was reborn, rejuvenated, at dawn. This model was also attributed, in a modified form, to other celestial bodies, notably the stars of the ecliptic, whose setting was understood as their descent into the primeval waters, from which they would be resurrected after seventy days of invisibility.⁴⁵ The other, and perhaps more relevant archetype, was Osiris, the god who had died and been returned to life as king of the underworld (*duat)⁴⁶ through the rites of mummification. It is a term associated with this deity, hasye,⁴⁷ “the blessed one”, which is used to describe the animal mummies in several of the magical texts, and which we will shortly investigate in greater detail.

12The essential purpose of these texts, then, seems to be the creation of a divine being, a creature with a link to the ritualist who created it, who could then be used in further rituals. The clearest example of this is in the recipe using the circaean falcon, in which, following the sacrifice, the ritualist speaks a formula, addressing the enshrined mummy as the sun-god, “who travels through the celestial vault and rises breathing fire”,⁴⁹ and then approaches the god “face-to-face”,⁵⁰ with the god becoming his paredros, or assistant, a spiritual being described in another text from the same papyrus as capable of carrying out almost limitless magical operations for the ritualist – cursing enemies, enchanting lovers, bringing food and riches, healing disease, revealing secret knowledge, and even guaranteeing salvation after death.⁵¹

13Other rituals make narrower use of the hasye: one ritual in which a cat is drowned calls upon the god to curse the ritualist’s enemies, although a pendant to the text adds that it can also be used to bind others in love or send dreams.⁵² This last use – the sending of dreams – occurs in two other examples, in which the animal mummies, a scarab and an animal whose name is lost, are called upon to appear in a vision to a third party.⁵³ Dream-sending,⁵⁴ in those texts where the purpose is most explicit, often seems to be type of love spell: the deity is invoked to appear to someone sleeping, often in the form of “the god that they worship”,⁵⁵ and instruct them to fulfil the desires of whoever commissioned the spell.

14In many cases, however, the hasye is not addressed directly as a god; in several, the bodies of the smaller divinised animals may be worn – in whole or in part, bound to the body of the ritualist or held in a ring – in various kinds of spells: a spell of ritual ascent in which the practitioner rises to the heavens to gaze upon the sun god,⁵⁶ a bowl-divination ritual in which the amulet may be bound to the boy-medium or the bowl itself,⁵⁷ and in love spells.⁵⁸ Each of these relies on the power of the divine animal – to allow the user to see the gods or to call them into his bowl of water and oil, or to attract a lover in a similar fashion – but the mummy is, in most cases, not directly invoked as a divine agent;⁵⁹ their power seems to act automatically.

15In still other cases, the mummy may not be used at all, although it may still be buried;⁶⁰ instead, the asphyxiant is used, having apparently taken on the divine power of the hasye through the ritual which created it. There may be a parallel here in the use of the efflux (*redju)⁶¹ of a mummified wolf in a dream-sending ritual similar to those mentioned above, and this word would link the liquid with a larger constellation of concepts within the Egyptian religious system – the asphyxiant could be understood as taking on the fertilising power of the Nile, which was both the watery grave and the efflux of Osiris, and the life-giving substance for which the liquids used in libations (milk, oil, water) were representatives.⁶²

16We have already seen that materia from the ritualist might be placed inside the bandages to create a link between human and animal-god, but a link in the other direction might be created by consuming part of the asphyxiant; in the ritual with the circaean falcon the practitioner is to drink the milk and honey in which the bird just died before sunrise (the time of the sun’s rebirth), giving the ritualist a share in the hasye’s divinity, a “part of divine inspiration” as the text puts it.⁶³ In other rituals, this “divine inspiration” is imparted by anointing the ritualist’s face,⁶⁴ or a ring that is to be worn⁶⁵ – both of these in rituals intended to bring about visions of the gods, in an ascent ritual and dream-oracle respectively.

17Again, however, the use of the asphyxiant may be more general; in two related love-spells the oil in which two fish are drowned is used to anoint the face and/or penis of a man before he sleeps with a woman to ensure her fidelity,⁶⁶ and in other cases the asphyxiant of various animals can be used to cure drunks and splenetics,⁶⁷ and ease child-birth.⁶⁸ The asphyxiant’s power is not limited to positive effects; in other texts the liquid can be fed to individuals to, for example, inflict them with sores,⁶⁹ or even kill them.⁷⁰ The asphyxiant from the same animal may even have very different purposes: the honey and milk in which the circaean falcon was drowned imparts divinity, but in another spell in which a falcon is drowned in wine, the asphyxiant kills a man who drinks it,⁷¹ while the wine in which a scorpion-fish is drowned may cure splenetics, but cause a woman to bleed uncontrollably; the only cure is for her to eat the cooked body of the fish itself.⁷²

18This dual function of the asphyxiant highlights an idea which will recur in this discussion: the position of the living human relative to the animal and god. If the same liquid, and the same animal, may have quite different effects, then the difference between killing and imparting divine power cannot be due to either of these variables. Instead, it seems that the efflux of the hasye takes on a general numinous power, which may either heal or harm depending on the ritual, and in some cases, the individual who drinks it: it might inspire an initiated ritualist with whom the god shares a special relationship, but kill an uninitiated person who should not be drinking the divine liquid. Like semen – another archetypal liquid which could be life-giving in some contexts, but poison in others – the efflux of the hasye could be dangerous when out of place.⁷³

19We have passed over a use of the hasye that is somewhat shocking: the eating of it, cooked, by a woman to cure bleeding induced by its asphyxiant, and it is this last use, that of the body of the drowned animal as ritual materia, that I will consider now. In one text a shrew and two lunar beetles are drowned and then pounded and mixed with other materia in an offering to the moon goddess in a love spell,⁷⁴ while in another the pounded body of half of a beetle is mixed with other materia and fed to a woman in a love spell⁷⁵ – the other half is worn as an amulet, creating a similar link between man and woman as that between animal-mummy and practitioner used in other spells. This consumption, reminiscent of the drinking of the asphyxiant in other rituals, is carried on in a spell in which a mole, an animal believed to have prophetic powers, is cooked in the water in which it was drowned along with other materia, and consumed at sunrise to give foreknowledge of the following day;⁷⁶ here we can see an echo of an idea, reported by the Neoplatonist Porphyry, that certain animals – such as ravens, moles and falcons – had a prophetic power in life, which resided in their body after death, and could be imparted to a human who consumed their body parts.⁷⁷

20Alongside consumption, another recipe uses the body of a drowned circaean falcon to create a phylactery to grant this same foreknowledge, binding the eyes and tongue, and other materia, with the falcon’s skin, and wearing it around the neck with its tendons as the cord.⁷⁸ In still other cases, the bodies of the hasye, in whole or in part, could be used to create unguents to treat problems as diverse as intestinal worms and gout.⁷⁹

21In studying these rituals, certain patterns in the use of animals seem to become apparent: there are recurrent types of animals – falcons, scarab beetles, shrews, fishes – as well as certain others – swallows, hoopoes, lizards – which appear less often. While none of the animals would have been particularly rare in Egypt, neither are any of them among the usual domestic creatures used for sacrifice (chickens, sheep, cows), and both of these facts – the recurrence of certain animals, and the appearance of other less usual species, may be significant. The sacredness of certain animals to certain gods – Greek, but in particular, Egyptian – was a key aspect of the cults of Roman Egypt, and this is clearly an important factor behind the choice of animal in several rituals.⁸⁰ The ritual in which a circaean falcon is divinised and made into a paredros includes a formula calling upon the god and constellation Orion, by both name and epithets, as well as descriptions – “the one who is young in the early morning and old in the evening” – which imply a solar character.⁸¹ As we shall shortly see, the animal mummies of the earlier Egyptian cults were believed to be sacred to particular deities, and to become, upon their deaths, and in some cases during in their lives, incarnations of these gods. As we have noted, the circaean falcon, perhaps the peregrine,⁸² or another species identified with the sacred Horus-falcon, was presumably understood not simply as being divinised, but as becoming an incarnation of the solar-stellar-deity.

22A similar, though more complex, link to a god is found in the case of the curse calling upon the spirit of the drowned cat, addressed variously as a “cat-faced god”,⁸³ “Seth-Typhon”,⁸⁴ the “steersman of the sacred boat”,⁸⁵ and the “form of Helios”,⁸⁶ the Greek name of the sun-god. The spell draws upon the solar associations of the god Seth in his role as the defender of the solar-barque,⁸⁷ and seems to use this link to connect him not with his usual animal, the donkey, but with the smaller – and hence more manageable – cat, one of the key solar animals.⁸⁸

23In other cases, the links may be clearer: the scarab beetle is a recurrent symbol of the reborn sun, and thus we find scarab beetles specified as “solar” used as hasye in a solar ritual, but by describing them as “lunar” or “of Mars” they may be linked to other heavenly bodies⁹⁰. Similarly, the shrew, a symbol of the night-sun, had lunar associations – the moon played the role of the sun by night, and reflected its light.⁹¹ In a few other cases, the links are less direct, but can be guessed: the Nile Perch, a fish sacred at Esna to the goddess Neith,⁹² is used in a love-spell calling upon a syncretistic goddess,⁹³ while the lizard, a being both believed to carry disease and often used in medicine, is employed in a ritual linked to the healing god Imhotep-Asclepius.⁹⁴

24In some instances, the link to a god is less certain; the association of the shrew with the blind eye of the night sun might explain its use in blinding spell,⁹⁵ and the foresight of the mole and the falcon, as we have seen, is the reason for their use in spells for foreknowledge.⁹⁶ Other cases are still harder to interpret, and the linking of swallows to incurable sores or red-mullets to childbirth may be either arbitrary, or linked to zoological, rather than religious, conceptions of these animals.⁹⁷ In many of these cases, in particular those in the Cyranides, there is little to suggest that the animal is considered a god, and we begin to get a sense that drowning was simply considered a particularly efficacious manner of killing an animal whose materia is required.⁹⁸ These spells often have no description of mummification, and no spells spoken to the animal before or after drowning. This suggests a development, albeit one which must be inferred, since it is not clear from the attestations in the chronologically-limited range of texts which survive.

25Perhaps the earliest rituals, inspired by the kinds of animal cults which we will consider later, involved the creation of hasye which were considered as gods, and used in rituals comparable to the paredros, dream-sending, and curse spells we have discussed. From this original function, other uses may have developed – the wearing of the animal as an amulet, or the burial of the mummy and the use of the asphyxiant in its place. Finally, the understanding of the drowned animal as a particular kind of divine being may have become increasingly restricted, and replaced by a conception that killing it in this way was simply a way to produce efficacious materia, either the asphyxiant, or parts of the hasye’s own body, to be used in producing incense or even unguents.

26The clearest indication of this development is the fact that the specific term linked to these beings, hasye, comes from Egyptian, and is linked to a religious concept; while ritual texts from other traditions may require the killing of animals in specific ways, drowning seems almost entirely restricted to Graeco-Egyptian texts, and does not seem to be found in material produced after the collapse of the Egyptian temples in the 3rd and 4th centuries, and the disappearance of traditional Egyptian religious practices which accompanied it.⁹⁹

27Nonetheless, even if the creation of the hasye makes them unique, they share the underlying logic of their use with other objects used in Graeco-Egyptian magical rituals.¹⁰⁰ This is particularly apparent in the dream-sending rituals, and other love spells; as we have seen, the request may be spoken to the hasye or written down and placed either beside it or in its mouth. Similar usage is made of figurines of the gods Selene and Eros (which have spells spoken to them or placed inside their bodies), figurines of animals – a wax hippopotamus or a clay jackal anointed with milk and the efflux of a wolf mummy.¹⁰¹ In each of these cases, the statue seems to serve, like the animal mummy, as the physical body of a divine spirit, a focal point which allows the human ritualist to interact with it. Even closer are rituals where the dead – human or animal – may be used for the same purposes, with other spells instructing the ritualist to place written formulae in the mouths of human mummies, dead dogs, or black cats who are biaiothanatoi (“violently-dead”);¹⁰² this last term refers to the belief within the Hellenic tradition that those who had died violently, became a special category of dead, empowered by their wrath and bound to restlessly wander the earth, and the same was true of the aōroi, those who died before their time – unmarried or childless.¹⁰³ While their bodies served a similar function to the bodies of the hasye, and to statues of animals or gods, they are understood as a specific category of the dead, with a similar (though distinct) cultural importance to the Egyptian concept of the hasye. It is to this concept that we will now turn.

The river and the white crocodile

28The term hasye, “blessed one”,¹⁰⁴ derives from the verb hōs,¹⁰⁵ “to bless, praise”. The word is attested very early in the written history of Egyptian, and we can find the deceased referred to as the “blessed one” of his father, his mother, the king, or the people as early as the Old Kingdom, where “blessed one” is often accompanied by or replaced by “beloved one”, that is, both are used to describe individuals who are loved and praised in both life and death.¹⁰⁶ This more general sense never seems to have entirely disappeared, and it was used, as late as the Graeco-Roman period, in some contexts even for the living, who might be particularly “blessed” by their divine or royal superiors.¹⁰⁷ Nonetheless, the title seems to have been strongly associated with the dead, and by at least the New Kingdom it is used to refer to the justified dead, originally perhaps linked to their attainment of immortality through the “blessings” of the king and gods.¹⁰⁸ A still more specific usage seems to have developed, perhaps around the 30th Dynasty (380-343 BCE), in which it referred to a particular sub-category of the justified dead, of which Osiris might be one;¹⁰⁹ in certain documents these are distinguished from the dead who are simply “people” (*remetju),¹¹⁰ and these same hasye are often referred to as hri, “exalted one”, apparently a title of respect.¹¹¹

29Texts from between the 30th Dynasty and the Roman Period, and found on funerary temples, stelae, statues, mummy labels, and in graffiti, attest to the title, either in its Egyptian form,¹¹² hasye, transliterated into Greek as hasiēs or esiēs,¹¹³ or into Aramaic as haṣie.¹¹⁴ While all justified dead were accorded the title “Osiris” – and more rarely, “Hathor”, for women¹¹⁵ – the number given the titles hasye and hri is far smaller. The most important of these seem to have been Petese and Pihor, two brothers who were worshipped in a small temple at Dendur in Nubia, built c. 15 BCE, and in the Kharga oasis in the Western Desert. Although little is known of their lives, they are depicted as deified humans in the temple carvings, Petese wearing the *atef-crown of Osiris, and both wearing crowns, solar-disks draped with guardian serpents, and lion-tails indicating their divinisation; they are referred to as hasye both in the temple inscriptions and in graffiti left by their worshippers.¹¹⁶ As is the case with many other categories of divine being in Roman Egypt, the word hasye was also used in theophoric personal names, such as Taphasiēs (“the one of the hasye”).¹¹⁷

30An important group of sources for the hasye are the legal documents from Ptolemaic Thebes belonging to the hereditary choachytes, individuals whose employment it was to carry out funerary liturgies for the deceased in the tombs of the west bank¹¹⁸. These liturgies, for which they received payment, were the possession of individual choachytes, and could, in consequence, be transferred in legal documents. Several of these deeds of transfer of liturgies survive, in both Greek and Demotic copies. Generally, they list the individuals who are interred in each tomb, either individually, or by the family or group they belong to, and there is repeated mention of individuals as hasye, in Greek either transliterated as hasiēs or translated as hupobrokhios, literally “submerged one”.¹¹⁹ Pestman notes that the hasye are often mentioned as “single persons”, buried without the relatives usually mentioned in the context of “normal persons”.¹²⁰

31The clearest description of the process of becoming a hasye comes from the Demotic literary text Setna I, dating to the Ptolemaic period. In this tale, Prince Naneferkaptah has stolen the Book of Thoth in an act of hubris, and in revenge the god causes the prince and his family to die, one by one. While travelling down the Nile on the royal yacht, first his son, Merib, then his wife, Ihweret, and finally he himself, fall into the water, “become hasye”, and are embalmed as hri.¹²¹

32While none of these sources are explicit, they collectively suggest that, at least by the 4th century BCE, the term hasye had developed into a title for a specific category of the dead, and that, as in the case of the animals used in the later magical papyri, this title was linked to being drowned. This hypothesis has a long history,¹²² and is supported by the statements of two Greek-language authors who would have been contemporaries of the hasye-cults – Herodotus and Aelian. In the section of the Histories dealing with Egyptian customs, Herodotus tells us that

Anyone, Egyptian or foreigner alike, known to have been carried off by a crocodile or drowned by the river itself, must by all means be embalmed and wrapped as attractively as possible and buried in a sacred coffin by the people of the place where he is cast ashore; none of his relatives or friends may touch him, but his body is considered something more than human, and is handled and buried by the priests of the Nile themselves.¹²³

33Aelian, who probably had never visited Egypt, provides an echo of this when he tells us that the people from the city of Ombos are proud when their children are eaten by crocodiles, “having produced food and a meal for a god”.¹²⁴ While the concept described here – including those whose bodies have been washed ashore having been killed by crocodiles – is slightly broader than the idea of “drowned”, we can see clear overlaps – many of those killed by crocodiles would drown before being eaten, and many of those who drowned might have their bodies subsequently eaten by crocodiles and other aquatic animals, and in both cases what remained of their bodies would be found in and retrieved from the Nile waters. An interesting link between these literary texts and the Ptolemaic evidence may be found in the references in the liturgies to some of the hasye as “belonging” to certain individuals – perhaps indicating that these are the “people of the place” mentioned by Herodotus who arranged for their burial as a pious act.¹²⁵

34The concept of the drowned hasye seems to be a fairly late development in Egyptian theology; in earlier periods drowning, like being burned alive, was seen as a form of death that damned the individual, since their body might be lost, damaged or destroyed, preventing them from receiving the funerary rites generally understood to confer post-mortem immortality.¹²⁶ The development of the idea of drowning as a particularly sacred form of death may be in part linked to a lessening of the importance of the funerary rites; in the sequel to the tale mentioned above, Setna II, the protagonist witnesses a poor man buried in a carpet whose good deeds led to Osiris transferring to him the funerary equipment of a rich man, who is perpetually punished by having his eye used as a door-hinge in the underworld.¹²⁷ While this account does not discount the importance of funeral rites, it stresses that blessed immortality is ultimately granted by Osiris, and not simply by the funeral rites in themselves.¹²⁸

35But a more concrete rationale seems to be found in identification with Osiris; as we have seen, it was usual for the dead to become “an Osiris”, but death in the water seems to have brought about a more complete identification. Egyptian sources are famously laconic about the death of Osiris; the fullest accounts come from Greek sources. Plutarch describes how Osiris was killed by being sealed in a casket by his brother Seth-Typhon, which was then cast into the Nile, before being later hacked into many pieces by Seth after Isis recovered the casket.¹²⁹ Here, it is unclear if he dies through suffocation or drowning, or even if his manner of death is important to the author. Diodorus Siculus, offering a somewhat syncretised explanation, describes him as being torn apart by Titans in an echo of the death of Dionysus-Zagreus, with whom Osiris was often identified by Greeks.¹³⁰

36By contrast, Egyptian sources rarely mention the death itself; their focus is on the finding of his body, his funerary rites and resurrection. A few texts are more explicit; several describe Seth throwing him down on the earth,¹³¹ but more common are descriptions where Osiris is *mehi, an ambiguous verb which may mean “to swim, to irrigate”, “to launch” a boat, “to float”, for a river “to flood”, and perhaps even “to drown”.¹³² Seth is said to have caused Osiris to be “mehi in the waters of the underworld, every limb separated”,¹³³ while Horus, Isis and Nephthys find Osiris “in the place where he had mehi”, and pull him out of the water.¹³⁴ The sense is ambiguous here: did he drown, or was he killed and dismembered on land before being cast into the water?¹³⁵ The opinion of Francis Llewellyn Griffith, who first drew the pieces of evidence I have listed here into a coherent model, was that, in being drowned the water, the hasye re-enacted the death of Osiris, and thus became assimilated to him.¹³⁶ This model has been predominant in studies of late Egyptian religion for at least eighty years;¹³⁷ the earliest of these studies drew heavily upon the Roman-era magical rituals describing the drowning of animals, but also made reference to James Frazer’s Golden Bough, which discussed the killing of the sacred king, in this case by drowning, as a central act of early religion. It was often suggested that the death by drowning of Osiris was a memory of an early ritual in which the kings of Egypt were symbolically killed, although this detail is usually disregarded in more recent work.¹³⁸ An important counter-hypothesis was proposed by Jan Quaegebeur in a series of articles in the 1970s, and it is worth addressing his objections in detail.

37Quaegebeur’s objection to the idea that the hasye were the drowned and blessed dead rested on several points:¹³⁹

1. He noted that the etymologically original meaning of hasye was “the blessed one”, so that even if it referred in the late Roman period to someone who had been immersed, or even drowned, this was not necessarily true of earlier periods.¹⁴⁰
2. In the funerary papyrus P. Rhind the term *ma’atiu (“justified”) is used in Hieratic to refer to the justified dead in general, but in the parallel Demotic version the term hasye is used, implying that the two terms were equivalent, and referred simply to the blessed dead.¹⁴¹
3. He argued that the name “Moses” was understood by Egyptians as being derived from mw-ḥsy “water-hasye”, and since Moses was saved from the water, and not drowned, the term could not refer to drowning.¹⁴²
4. He argued that Osiris did not drown in the Egyptian myth, and so hasye, applied to him, could not mean “drowned”.¹⁴³
5. He noted that the Graeco-Egyptian term Hereseion/Thareseion seemed to refer to a temple of a somewhat mysterious deity known as Haresies, which he understood as rendering “Horus-hasye”; he linked this name to a myth told by Diodorus in which Horus is killed by the Titans, and found dead in the water by his mother Isis, who resurrects him.¹⁴⁴ Quaegebeur traced this story back in turn to an episode from the Contest of Horus and Seth in which Horus hides under the water in the form of a hippopotamus,¹⁴⁵ and suggests that Diodorus misunderstood this story of hiding in water as a story of drowning.¹⁴⁶
6. He noted the mention of the hasye as a type of malevolent spirit, arguing that in this case the term referred simply to a type of aquatic demon.¹⁴⁷
7. Finally, he noted that the number of hasye in the legal texts of the choachytes is rather large, and more than would be expected if it referred to individuals who had been drowned.¹⁴⁸

38Based on the force of these objections, Quaegebeur argued that hasye referred instead to the beneficiaries of a ritual of immersion carried out on the dead, a kind of “post-mortem baptism” which was intended to parallel the immersion of Osiris in the water after his death;¹⁴⁹ he compared this, somewhat infelicitously, to the immersion of the cat in the magical ritual discussed above,¹⁵⁰ and to the ritual bath described by Apuleius as taking place during the initiation into the Isis cult in his Metamorphoses.¹⁵¹ Claude Traunecker, building upon this idea, has suggested that the status of hasye was constituted by receiving libations of water, wine, milk, and beer every ten days, paralleling the libations offered to Osiris. ¹⁵²

39Before engaging with his points in detail, it is worth pointing out the greatest problem with this hypothesis: there is no evidence for a ritual immersion given to mummies. Early studies of mummification often focused on the idea that the process of desiccating the body with natron was carried out in a natron solution, and one modern recreation has used this method, but the idea is generally agreed to rest upon a misreading of Herodotus; no bath permitting an immersion – either ritual or practical – has been found among the embalming assemblages discovered in Egypt, and studies of mummies show results generally consistent with dry natron application.¹⁵³ Similarly, while some scholars have suggested that the ritual bath undergone by the Isaiac initiate may have been linked to the death and resurrection of Osiris, there is no clear suggestion of this in the text itself, no reason to think the Isaiac initiation procedure derived from the treatment of mummies, and a stronger argument that this represents simply a case of ritual purification.¹⁵⁴ Finally, it is very clear that the cat used in the magical curse was not simply submerged, but drowned.

40Let us now engage with his points one by one. His first point – that the hasye as a “justified drowned individual” cannot be the original sense of the word – is necessarily true. As I have discussed above, the term seems to have emerged very early in Egyptian history, and to have developed multiple concurrent and context dependent meanings. Based on the testimony of Herodotus and the examples assembled by Alan Rowe we might assume that the concept of deification by death in water was present by at least the 5th century BCE, and connected with the term hasye by at least the 4th.¹⁵⁵

41This polysemy may account for the usage of the Rhind papyrus, which translates ma’ati (“justified”) as hasye. It seems clear that in certain contexts hasye referred not to the justified in general, but rather to a specific category of dead, as Quaegebeur himself admits. The ma’ati-hasye translation is probably best understood therefore as using the older, more general sense of hasye.¹⁵⁶

42As we have seen, the evidence regarding the manner of Osiris’ death is unclear, but I think there is a solution which preserves Griffith’s brilliant synthesis of the evidence with Quaegebeur’s objection. All of the sources agree that his body was found and retrieved by his sister-wife Isis in the waters of the Nile,¹⁵⁷ an act which seems to have been re-enacted at the annual Wag-festival,¹⁵⁸ and this may thus have been a more important fact to his worshippers than the precise manner of his death. This could be the link between Osiris and the hasye, who would, like him, have been found floating dead in the Nile, and pulled out of the water. It is likely that those who died in water (by whatever means) were generally discovered considerably downstream, and that they would often be buried by individuals who did not know who they were, and who thus referred to them simply as hasye, a title of respect for the justified dead; this is consistent with the evidence from the archive of the Choachytes, where the hasye are often anonymous and regularly lack the filiations and occupations which are given for the other dead.¹⁵⁹ This may have led first to a secondary meaning, best attested in this archive, in which hasye (in reference to a specific individual) was a way of designating that they had undergone the process of death and discovery in water which led to the blessed status described by Herodotus. Later, hasye may have taken on the still more specific sense of “a drowned individual”, suggested by Setna I and the Roman magical texts, and preserved in Coptic, as drowning would be the most common (if not only) means by which individuals found in the water would have died.

43This explanation would also seem to help us to understand the case of Quaegebeur’s derivation of Moses from “water-hasye”, based on a passage from Josephus in which the Judaean writer attempts to understand the name in terms of contemporary Egyptian. It seems likely, if uncertain, that the element which Josephus gives as -esēs derives from hasye, though it is unclear that this interpretation originated from a native-language Egyptian-speaker, rather than a non-speaker who may have misunderstood or misrepresented the word or its meaning. ¹⁶⁰Černý points out that another author who discussed the name, Clement of Alexandria, added that “some [Egyptians] call ‘Moses’ one who has died in water”,¹⁶¹ suggesting that both Josephus and Clement are in (confused) agreement with the idea that a hasye was prototypically an individual retrieved from the water, even if Moses himself was a rare instance of a living hasye.

44Quaegebeur’s discussion of Horus-hasye is difficult to assess due to lack of evidence. This figure does not seem to be attested in any Egyptian-language texts, and so this interpretation of his name rests on Greek transliterations which obscure the underlying Egyptian morphemes; the element “Horus” is certain, the element hasye less so, although it seems to be the best candidate.¹⁶² More problematic is his reading of the myth: to suggest that a Roman-period telling of a myth derives from a specific New Kingdom account of a related cycle is problematic: the versions that survive are only single variants of traditions that were, no doubt, much larger, and it is probable that there were many other episodes which do not survive. Diodorus’ story in particular seems to attribute (again) the death of Dionysus-Zagreus to another deity, this time Horus – a child killed by Titans and resurrected by his parent – and has echoes of magical historiolae in which Isis saves her son from the attacks of enemies or disease,¹⁶³ and may thus represent a late, syncretistic addition to the tradition. There is no clear reason to think that Diodorus is significantly misrepresenting this myth, and the story explicitly states that Horus was (like the hasye) found dead in the water, even if it does not tell us how he was killed.

45Let us turn to his final two points. As Quaegebeur notes, hasye are mentioned in the texts known as self-dedications; these consist of contracts, written in the early 2nd century BCE, in which suppliants agree to become servants of deities in return for supernatural protection, and hasye are listed among the various hostile powers and diseases from which the deity will protect their servants.¹⁶⁴ Quaegebeur’s implied argument is that the hasye here are demons who live beneath the water, and can thus be understood as in some sense “submerged” or “immersed”. But the category of hostile powers in the Egyptian worldview was very large, and included not only various categories of non-human spirits, but also the souls of the dead and a category of being known as *akh, a word whose primary referent seems to be to individuals who had undergone the funerary rites which transformed them into immortal spirits; this is the same term used for “ghosts” in stories in which the living encounter the dead in their tombs,¹⁶⁵ and they are mentioned alongside the hasye in the self-dedications. It is thus far from clear that these texts prove that the hasye were aquatic demons rather than specific category of dead who might at times threaten the living.

46His final point – that there are too many hasye in the choachyte archives – is more difficult to address. It is true that in reading the lists of individuals in liturgy sales one gets the impression that there are a great many hasye, although the vague references to “families” and other plural groups in the case of non-hasye makes it very difficult to get a sense of the precise number of individuals being referred to. If we assume that each reference to a plural refers to two individuals (a number almost certainly far too low), then perhaps one in ten mummies (11 %) seems to be a hasye; more reasonable assumptions, counting plurals as five or ten individuals, would yield numbers closer to one in twenty (4-7 %).¹⁶⁶ This is still more than we might expect to have drowned, and certainly more than we find in modern Egypt.¹⁶⁷ There are a few explanations we could put forward for this: it may be, for example, that since the drowned were particularly sacred they were more likely to be included in liturgies, which were often paid for by others,¹⁶⁸ than those who died in other ways (who might be, for example, too poor to pay for funerary rites). It may also be that their liturgies continued for longer than other individuals; even if particular hasye did not have temples built in their honour, as did Petese and Pihor, the level of worship and reverence they inspired may have led to their liturgies being paid for over spans of time in which individuals who did not drown would be forgotten by their descendants, or even have their lines of descent disappear; this would result in a multiplier effect with each generation.¹⁶⁹ Nonetheless, an interesting piece of cross-cultural research may suggest that deaths in water – including drownings, but also deaths by crocodile and so on – were much more common in pre-modern times: a study of coroner’s reports from Tudor England has found that drownings accounted for half of all accidental deaths, twenty-five times more than the comparable figure (2 %) for England in 2010.¹⁷⁰ People at this time spent much more time in and around lakes and rivers – travelling by boat, swimming, washing themselves and their clothes, relieving themselves – all of which made their drowning much more likely. None of this is decisive, but it does suggest that the apparently large number of hasye is not necessarily inconsistent with the idea that they were drowned, or otherwise found dead in the water, especially in the absence of clearly attested alternatives.

47In sum, Quaegebeur’s ideas should make us hesitant to translate every occurrence of hasye – especially in pre-Ptolemaic texts – straightforwardly as “drowned”, as Rowe did in his study.¹⁷¹ Instead, I have suggested that the term’s meaning is heavily dependent upon context, and could refer to (1) an individual (living or dead) “blessed” by another, (2) the justified dead in general, (3) dead recovered from the Nile in particular, and (4) even more specifically, those who had drowned. This context-sensitive meaning may seem overly complex, but does account well for the evidence, and it is a general feature of human language. We might note a relevant parallel here in the Greek word martus, which, by the end of the 1st century, could mean either “witness” in a general, religious, or specifically legal sense, a planet in a particular astrological aspect, or “one who had died for their (Christian) faith”, with the meaning dependent upon, and clear from, the context.¹⁷²

48If the meaning of the term prior to the Roman-era magical texts is ambiguous, a survey of the usage of hasye, and parallel terms, in these texts seems to demonstrate a much more straightforward usage. A representative sample of these can be found in table 2 below.

49What is immediately obvious in these texts is the highly standard language; the instructions generally begin “take (X animal)”, followed by a verb describing what is to be done to the animal, and often add the outcome. If we begin with the Demotic examples, we can see that two terms are used are “make [the animal] hasye”, and the more periphrastic “make [the animal] go hasye”. In examples (8 & 10) we can see that one “makes [an animal] hasye” in a liquid – the milk of a black cow, and in example (9) we can see that “putting into” a liquid seems to have a similar sense, although the text adds “make [the animal] hasye” to give a more explicit sense of what “putting into” entails, or results in. This periphrastic construction – “make hasye” seems to be that which lies behind example (5), where the Greek can be understood as an exact calque of the Demotic: “make it hasye, casting the [cat’s] body into the water”; what this entails is again made explicit shortly thereafter, when the process is described as a “drowning”.

50The phrase “go hasye” is used in Coptic to refer simply to “drowning”, apparently without the older religious sense, and the verbs used (“do/become”, “go”) are the same as those found in the Demotic examples above.¹⁷³ We may see that, in these examples, “go hasye” is the equivalent of intransitive “drown” (“he drowned”), while the transitive (“he drowned a cat”) may be expressed with either one or two auxiliary verbs “make hasye” (“make X into a drowned person”), or “make X go hasye” (“make X drown”).

51While (5) seems to offer a calque of the Egyptian-language expressions, the majority of the Greek texts do not use a calque; instead, examples (1, 4, 6 & 7) describe the process as “deification”. As before, we may note that we find the verb “deify” used in exactly the same positions as “make (go) hasye” – the ritualist is to take an animal, “deify” it in a liquid, and this process is associated with “putting it in” a liquid (1, 7) and “drowning” it (1). I would suggest that “deify” should be understood as directly translation of “make hasye”; lacking a specific noun corresponding to hasye, Greeks could either borrow the Egyptian word to form a calque, or (the more common choice) translate the concept – hasye becomes “god”, and “make hasye” becomes “make (into) a god”, that is, “deify”.¹⁷⁴ As with “make hasye”, however, the phrase in this context referred not simply to deification in a generic sense, but more specifically to drowning. Example (4) describes deification both as a process (“deify/drown it”), and an outcome (“when it is deified/drowned…”), and this is even more explicit in example (1); the text begins by telling us to “submerge” the falcon until it is “deified”, and then adds, “that is, drown the falcon… until it is dead”. Here, “drowning” glosses “submerging” – a submerged animal would, of course, drown – but “deification” is glossed in the same way by “death”; that is, the “deification” refers precisely to death by drowning. In other texts from the Cyranides the verb “deify” is not used at all, and the instruction is simply to “drown” the animal (3). The equivalents between the Greek calques and translations are shown in table 3.

52As we have already noted, drowning in magical texts is often understood as a particularly marked, or efficacious, way to kill an animal, but this is not always directly linked to the concept of becoming a divine being. The same would seem to hold for the terminology itself; there is no doubt that, at least in these texts, hasye refers specifically to a being killed by drowning, and at times this is all it means.¹⁷⁵ We should remember that dying (and originally, being found dead) in water – of which drowning is the most prominent example – had an inherent, and inseparable meaning in late Egyptian religion – to drown was to become hasye, and a hasye was usually a being which had drowned – but in different contexts the religious aspects might come to the fore or recede, in the same way that a priest might be understood either as an individual with a special relationship to a god, or simply an individual with particular legal rights and obligations, depending on the situation.

53This evidence can be supplemented by the other appearances of the word hasye, in Egyptian or in transliterated Greek, in magical texts. It is often used as an epithet of Osiris, who is described in one text as the “good hasye of the hasye”,¹⁷⁶ that is, the “blessed one” of the “blessed ones”. One text describes how he “became hasye and was taken away by the river”,¹⁷⁷ where “becoming hasye” may be understood as a relatively explicit description of his drowning. In another, the ritualist threatens to reveal how Osiris “remained in the river three days and three nights, the hasye… belly eaten by fish”.¹⁷⁸ Hasye are also mentioned as a specific type of dead individual, with one text mentioning how the gods, in response to a spell, will release “all the hasye, all the unmarried-dead, all the wind-borne”;¹⁷⁹ here we see again the link between the untimely, restless dead and the hasye as special categories of the deceased. Likewise, spells which summon the gods to appear in divination rituals may also include the possibility of summoning the dead in general, and hasye in particular.¹⁸⁰ Finally, a spell to remove a bone stuck in the patient’s throat, invokes a “white crocodile, which is under the foam of the sea of flame, whose belly is full of the bones of every hasye”, commanding him to spit out one of these bones.¹⁸¹ This powerful image is reminiscent of the idea, stressed in Herodotus and Aelian, that death by crocodile was as efficacious as drowning in translating an individual into a hasye, but the specific mythic referent is perhaps to those hasye whose bodies have been lost in the Nile – prey to an archetypal divine crocodile – but who are nonetheless sanctified by their death. It is a reminder both of the divinity of the hasye, and of their fundamental status as dead, and by extension, vulnerable, beings.

54This picture of the hasye – a specifically Egyptian conception of an individual who had died in the water and taken on a particular divine status – clearly lies behind the animal mummies we have seen used in the rituals from Roman Egypt, but there is one element missing. There is no suggestion – apart from Frazer’s unsupported assumption of an original ritual involving the king – that any of the earlier hasye had been deliberately killed. For this aspect, we must turn to another expression of late Egyptian religion, the animal cults.

One god in one jar

55One of the most distinctive features of ancient Egyptian religion was the sacredness, and indeed divinity, of certain animals. This is a point noted by modern scholars, but it was also a stereotype held by their neighbours – Greek, Roman, Jewish, and Christian – and a particularity that seems to have been adopted as a marker of ethnic identity by the Egyptians themselves from the Late Period onwards.¹⁸² The oldest and most prominent animal cult was that of the Apis, a god incarnate in the form of a living bull, dating back to the Early Dynastic period,¹⁸³ but animal cults, and the mass-mummification of animals that accompanied them, seem to have become much more prominent around the 7th century BCE, and this heightened importance continued in the Ptolemaic and early Roman periods.¹⁸⁴ The quantities of animal mummies produced, interred, and subsequently excavated by archaeologists was enormous, to the extent that early excavations often discarded all but the most striking examples; hundreds of thousands of them were famously ground up, made into paper, used as fertiliser, or burned for fuel, both in Egypt, and in European countries, to which they were carried by the shipload.¹⁸⁵ A particularly famous cartoon from an 1890 edition of the British magazine Punch depicts a field fertilised by hundreds of cat mummies, who have risen, hasye-like, to terrify a group of English farmers (see fig. 3).

56Modern studies of animal mummies generally classify them according to four different categories, although it is unclear if Egyptians themselves would have made the same divisions.¹⁸⁶ The first category consists of the mummies of pets – individual companion animals which had particular sentimental value and were and mummified after their deaths by their owners; this practice is attested from the New Kingdom, but is relatively rare, and restricted to the highest echelons of society. The second category consists of victual mummies, animals, or their body parts, dressed and preserved as food to accompany the humans with whom they are buried. Attested from the 18th-20th Dynasties, this practice is clearly distinct from other mummification procedures, in that the preservation is intended to preserve their flesh as food rather than as the receptacle of an immortal spirit.

57Still rarer, though attested more-or-less continuously throughout Egyptian history, are the mummies of the god-animals, such as the Apis, Mnevis, and Buchis bulls, the rams of Khnum, and the crocodiles of Sobek. Each of these was an individual animal, worshipped as the incarnation of a particular deity, mummified after their deaths, and replaced by new living individuals of their same species, identified by particular physical characteristics, which represented the new incarnation of the same divine entity. This worship often extended to the families of these animals – the mothers of the Apis bulls were worshipped as incarnations of Isis, and mummified and interred in their own catacombs,¹⁸⁷ and even the children of the Apis enjoyed a cult.¹⁸⁸ Each of these categories was treated, both in life and in death, as in some way equivalent to human kings, who represented in their persons the incarnations of Horus; like the kings they lived in temples equivalent to the royal palace, and had funerary chapels dedicated to them.¹⁸⁹

58The most common category, however, is that known as the votive mummy, a type which appeared in the Late Period (c. 664-332 BCE), and is generally understood to have continued to be produced until the end of Egyptian cults in the Roman Period (c. 380 CE).¹⁹⁰ Many types of votive mummies are known – cats, dogs, fish, ibises, and falcons are the most common, although shrews, lizards, snakes, wolves and mongooses, as well as many other animals, are also known.¹⁹¹ These animals seem to have lived within the precincts of the temples to the gods to whom they were sacred – Sakhmet and Bast for cats, Anubis for dogs and wolves, Neith for the Nile Perch, Horus for falcons, Thoth for ibises and baboons, and so on – and to have been mummified and interred in catacombs associated with these temples in their deaths. It is these mummies which earlier generations used as fertiliser, and which can be found in many modern Egyptological collections. They were mummified in huge quantities – an estimated eight million dogs were buried at the Anubeion at Saqqara, four million ibises at the nearby ibis catacombs, nearly two million at the catacombs of Tuna el-Gebel.¹⁹² Hor of Sebennytos, a priest who lived in Saqqara in the 2nd century BCE, refers in one text to sixty-thousand ibises being kept there at one time,¹⁹³ a plausible number given the remains. Studies have estimated that, at the height of these cults, an average of forty-one ibises would need to have been mummified every day at Tuna el-Gebel,¹⁹⁴ and fifty dogs a day at Saqqara for over 400 years, to produce the numbers that survive.¹⁹⁵

59It was Jean Capart who first suggested, in a 1943 paper on cat mummies, that a temple industry on this scale could not have taken place without the deliberate killing of animals;¹⁹⁶ interestingly, this suggestion was inspired by the Roman-period magical text mentioning the drowning of a cat, and the idea of drowning has at times recurred in literature on animal cults.¹⁹⁷ This suggestion runs counter to the picture of the treatment of sacred animals which we inherit from Graeco-Roman authors, who stress the inviolability of their lives: Herodotus tells us that anyone who killed a sacred animal was put to death,¹⁹⁸ while Diodorus Siculus recounted the story of a Roman lynched by Egyptians for accidentally killing a cat, despite the attempts of royal officials to intervene.¹⁹⁹ It was a commonplace among Greek and Latin-language authors that certain animals were sacred in each region of Egypt, and that the locals did not kill or eat these animals, and were hostile to those that did, in particular those in neighbouring villages, who worshipped different species.²⁰⁰ To some extent this is likely an othering discourse, a product of the stereotype of Egyptians as excessively religious animal worshippers, but it is probably also true that killing a sacred animal was not generally permitted. A rare surviving Egyptian legal manual from the Ptolemaic period lists the maltreatment of sacred animals as a crime, an idea echoed, albeit with frustrating lack of detail, in a range of literary and documentary texts from Egypt.²⁰¹ Scholars who study the animal cults, notably Alain Charron, have therefore suggested that the killing of animals as votive mummies may have been the prerogative only of the priesthood, whose position entitled them to carry out acts that would have been sacrilegious for others, and that this act may have been hidden from the public.²⁰²

60The standard interpretation is that the votive mummies were bred en masse in specialised areas within the temple complexes.²⁰³ Hatcheries have been tentatively identified for crocodiles at Narmouthis and for ibises and falcons at Saqqara,²⁰⁴ along with enclosures for keeping them, and documentary papyri attest individuals whose titles identify them as attendants of these animals.²⁰⁵

61Despite norms against harming sacred animals, studies of the bodies of animal mummies suggests considerable maltreatment, as we might expect in a situation where animals were mass-bred to be killed. Six per cent of the dog skulls in the Teti cemetery at Saqqara showed that the dogs had suffered serious blunt force trauma during their lifetimes, and although the numbers were smaller for the Anubeion, these signs of trauma were still present in 0.75 % of the bones.²⁰⁶ This trauma indicates that many of these dogs had had their heads beaten badly enough to injure their bones, but not to kill them; presumably these beatings were used to discipline them, and we would expect the number of dogs who were beaten but did not have broken bones to be considerably larger. Similar evidence for mistreatment can be identified in other cases; an ibis showing signs of a fractured femur which had healed at an angle of ninety-degrees indicates an animal which had been fed, but which did not receive medical care for a serious injury.²⁰⁷

62If their lives were violent, their deaths were no less brutal; over the last decades Capart’s general hypothesis has been confirmed by imaging studies of hundreds of animal mummies, which have revealed death by head-trauma, strangulation indicated by displaced neck vertebrae, or, in the case of crocodiles, nostrils slit open.²⁰⁸ Even where there is no direct evidence of the cause of death, analysis of age distributions shows that the vast majority of cats and dogs died at less than one year of age: they were killed at very young ages.²⁰⁹ This does not seem to be because small or young animals were preferred, however; instead, they seem to have been killed when they had attained the minimum acceptable size, and a considerable amount of linen padding was often used to make them look larger.²¹⁰ The situation seems to have been similar in the case of fish and crocodiles, generally represented by small specimens.²¹¹ The picture that emerges is not that of the idyllic sacred animal menageries we might imagine, but rather of an industrial process reminiscent of modern industrial farms or pet mills. The methods of mummification used is similarly industrial; while a few examples show quite beautiful results, often with polychrome wrapping which is padded to reproduce the form of the living animal, more often they are bound cursorily in plain linen, often with flesh exposed, and apparently, in particular in the case of birds, after being simply dipped in resin, a process which may itself have been the means of killing them.²¹²

63As the term “votive mummies” implies, the reason for producing these animal mummies is usually understood to be the production of ex votos, divinised animals who could be paid for by visiting pilgrims before being interred in the associated cemeteries. In this way, they would function in an analogous manner to the small bronze votive statues of deities which were also wrapped in linen and placed in sacred spaces; both would serve as icons of the gods to which they were dedicated, and would serve both as symbols of piety, and as intercessors who might take messages and prayers to the gods to whom they were dedicated.²¹³

64This model has been problematised somewhat by more recent studies; there is little direct evidence for many of the key stages. DNA analysis of ibis mummies suggests that the ibises may have been wild flocks fed by the temple staff, in wild habitats or small local farms for short periods of time,²¹⁴ while work by Lidija McKnight and Stephanie Atherton-Woolham suggests that some pathologies – such as apparently broken necks – may in fact be artefacts of the mummification process.²¹⁵ Even the idea of animal mummies as votives is surprisingly lacking in evidence – the most explicit texts refer to the simultaneous burial of hundreds of animals, rather than dribs and drabs paid for by individual pilgrims.²¹⁶

65Nonetheless, the overall picture is hard to dismiss; the age profiles of dogs and cats, and the presence of head wounds, suggests that at least these animals were bred, killed, and mummified en masse; the killing may even have consisted simply in deliberately neglecting young animals who could not survive without the care of either humans or their parents.²¹⁷ The silence of Egyptian texts around the deaths of these animals, as with the death of Osiris, is almost total,²¹⁸ but as far as we can tell there is no evidence that they were drowned, as in the case of the later magical rituals, nor are they given the titles hasye or hri.²¹⁹

66Nonetheless, we do find a forerunner of the roles of the animal hasye in the cultic practices associated with these earlier mummies. However they had been treated in life, in death they had become gods, and they could be addressed as such. Several letters survive, calling upon the mummified ibises of Tuna el-Gebel for help with daily crises, invoking them as avatars of the god Thoth; these seem to have been placed in their catacombs or in the jars in which they were buried.²²⁰ Similarly, a petition to the mummified mother of the Apis has been found in Saqqara,²²¹ and a demotic tablet records a literary text in which an un-named king descends into the funerary chamber of the Apis bull and has a vision in a dream of a divine figure, presumably the Apis himself, who reproves his bad behaviour and instructs him how to make amends.²²² Both in their general outlines, and in their specifics, these prefigure the later magical rituals – the placement of written requests by the mummified bodies recalls the dream sending rituals, and the practice of revelation by incubation recalls the ritual in which the falcon is called upon to appear to the ritualist as his divine familiar.

67The most explicit evidence for the cultic importance of the animal mummies, however, comes from Hor of Sebennytos, the priest who mentioned the sixty-thousand sacred ibises of Saqqara. In the 2nd century BCE he seems to have lived in a small chapel near the entrance to the ibis catacombs, and to have worked as an oracle, answering questions for both himself and others through prophetic dreams, which he invoked through prayers whose form and contents are highly reminiscent of invocations in the later Demotic magical papyri.²²³ Two of his texts tell us that these dreams were, at least at times, received while he was within the animal cemeteries – in the tomb of the mummified Mnevis bull in one case, in the ibis catacombs in another.²²⁴ The record that we have preserved in Hor’s archive may be unique, but there is some suggestion that an analogous practice may have been carried out in the ibis and baboon catacombs at Tuna el-Gebel.²²⁵

68Hor’s close association with the ibis mummies, whom he invokes along with the other animal mummies in his spoken prayers²²⁶, may even have led him to identify with them. In several texts he records his concern for “sustenance in life and burial in death”, promised to him in dreams by Isis and Thoth, and these two terms are the same as those used in his writings to identify the treatment due to the ibises – sustenance and burial. ²²⁷The neglect of these two “rights” occasioned his conflicts with those in charge of the ibis cults, a conflict which lasted from at least 172 BCE to 147 BCE, and perhaps longer.²²⁸ These conflicts are recorded in surviving drafts of petitions, intended to ask the king and queen to intervene, as they apparently had on earlier occasions, to ensure the animal gods were treated fairly. Perhaps providing support for the picture of the mistreatment of animals sketched out earlier, several texts claim that the food of the ibises was stolen, that they died of hunger constantly, and that they were neglected night and day.²²⁹ But his concern for their post-death treatment seems to have been more acute than his concern for their care in life, and he complains in another text about the performance of a burial which was “bad in every way”.²³⁰ The earlier, apparently unsuccessful, reforms which his previous petitions had occasioned were apparently centred around ensuring that each ibis was buried in its own jar – “one god in one jar”, ²³¹as he evocatively puts it – in contrast to the widespread practice of interring multiple animals in a single jar, attested in many surviving mummies.²³² As an oracle, Hor spoke with the authority of the gods, and thus, in a sense, with the voice of the dead ibises complaining about their own mistreatment.²³³

69This highlights the underlying problem of the animal cults, and the contradiction which occasioned this paper: between the animal as a creature to be killed for human use, and the animal as a divine being whose help the human needs. Surprisingly, this is a contradiction which has been little discussed; Charron suggests that while “the animal counted”, it counted “much less than the mummy which was better able to be a divine receptacle”,²³⁴ while te Velde has suggested that what we see in the animal cults is “not killing life to destroy it, but to let it arise from death”. ²³⁵As I have already suggested, the “almost industrial”²³⁶ level of production implied by the number of animal mummies might point us towards a somewhat parallel phenomenon in the modern West, that of intensive animal (“factory”) farming. In both ancient and modern practice we see the rearing, killing, and processing of animals on almost incomprehensible scales – tens of thousands on the one hand, tens of billions on the other. The killing of juvenile animals whose mummies are made bulkier by padding even finds a parallel in the modern practice of developing fast-growing breeds which may be slaughtered at younger ages. Our understanding of these contradictory animal-human relations can be nuanced by the concept of “deading life”, proposed by the philosopher James Stanescu.²³⁷ By this jarring phrase Stanescu means “life whose production is fundamentally about its death, its consumption”.²³⁸ He uses it in reference to modern industrial farming, arguing that:

… factory farms are simply a manifestation of a classical question, one which even Cicero takes up. When reading Cicero on animals, we see that he replicates the common Stoic belief that all other animals exist for the purpose of being used by humans. Cicero would say of pigs: “What is the swine good for but to eat? whose life, Chrysippus says, was given it but as salt to keep it from putrefying”…²³⁹ In such a view… life is not living. Life is salt, life is refrigeration, life is merely a process and precursor to death. Life is but an adjunctment to the end product, death.²⁴⁰

70That is to say, the animals in modern industrial agriculture are not alive, they are merely not yet dead; their death is not their end, but the beginning of their existence – while they are alive, they are simply unfinished meat. Their lives have no value, save the value of their use.

71Modern industrial farming and ancient animal cults are not the same phenomena, but there are clear similarities in the incidental horrors they both produce, and the invisibility of the acts of killing which produce in one case, ready-made gods, and in the other, food; the goal of the votive animal cults was the production of animal mummies, and the raising and killing of animals were simply necessary steps in their production. As Stanescu puts it: “if animals are killed, it isn’t because there is a hatred for the animal, or a sacrificial moment, or anything of the sort. Rather, animals are killed as beings who are being produced for their death”.²⁴¹

72The magical papyri represent a sharp divergence from this, in that the killing, while still representing an instrumentalisation of the animal, is foregrounded as an act key to the function of the overall ritual. The animal must not simply die, but must be drowned, and must be drowned in a specific substance; the animal becomes Osiris, or, in the ritual of the circaean falcon mummy becomes a solar Orion, who dies in the water at the edge of the world and rises reborn in the morning. Each element may be understood in mythic terms – the animal is the dying and resurrected god, and the asphyxiant is the primeval Nile water – but the ritualist, who kills, remains invisible. In the Osiris myth, the killer is present as Seth, who must be punished for his crime,²⁴² but the practitioner never takes this place in the ritual scene. His role as killer, like that of the butcher-priests of the votive animal cults, is firmly un-mythologised. It is to this final question – the relationship between killer and hasye, that we will now turn.

The death and resurrection of the beloved beast

73We have seen that the practice of creating magical assistants and materia through drowning animals seems to have arisen from two Egyptian religious practices, the veneration of hasye and the production of votive mummies in the animal cults, but the magical texts may not represent the only place where these two elements have become intertwined. First recorded in the writings of the 1st-century CE Latin author Pliny the Elder, there is a minor but persistent tradition among Graeco-Roman authors that the Apis bull, the foremost of the animal gods of Egypt, was ritually drowned.

74“It is not lawful for [the Apis] to exceed a certain number of years of life,” he tells us, “and they kill it by drowning it in the fountain of the priests, proceeding with lamentation to look for another to put in its place, and they go on in mourning till they have found one, actually shaving the hair off their heads”.²⁴³ While Pliny provides us with the earliest attestation of this idea, it is repeated by several other authors,²⁴⁴ notably Plutarch, who sets the lifespan of the Apis at 25 years,²⁴⁵ the square of the sacred number five, and the 4th-century author Ammianus Marcellinus, who adds that a specially marked cow was killed and offered to him, presumably as a mate.²⁴⁶

75Alexandre Moret, the author who discussed this theme at greatest length in the 1920s, used these texts to argue that the death of the Apis was linked to the death of the sacred king theorised by Frazer;²⁴⁷ the Apis served as a substitute for the king, drowning in the place of the human pharaoh, who in turn represented the drowned god Osiris. Again, however, there is little archaeological or Egyptian-language evidence to substantiate these ideas; the fact that no Egyptian source makes mention of the killing of the Apis is not decisive proof against the practice, since we have seen a similar silence about the death of animals intended as votive mummies, and Osiris himself. More problematic is the fact that the dates of the deaths of many Apis bulls are known – and for these their ages can be calculated, since each was born shortly after the death of his predecessor. None of them died at 25 years of age; two lived longer than this, but most shorter, which is perhaps unsurprising given that the natural lifespan of domestic cattle is around 20 years.²⁴⁸

76How do we resolve this problem?²⁴⁹ It is of course possible that the bulls were still drowned, but simply at different ages, although, in the absence of a rationale for the different ages at which they were killed, this seems unlikely. Jean Vercoutter suggested that the Graeco-Roman accounts derive from a misunderstanding of a real practice, a ritual bath which took place before or after death.²⁵⁰ But I think there is another possibility, speculative though it may be: Pliny’s writings may be approximately contemporary with the earliest stages of the writing of the Cyranides, which discusses the production of magical hasye, a period by which the translation “deify” had already developed for “make hasye”. This, and the fact that the examples in the Cyranides have already lost their specifically religious connotations, suggest that the practice had already emerged, and been known for some time by this date.²⁵¹ It seems likely that the making of animal hasye developed out of the animal cults: we might imagine individual priests producing more specialised divine assistants for their own purposes or for private clients, and an imprecise knowledge of these practices, transmitted in late Hellenistic magical texts, may have led Pliny or his sources to attribute the practice of drowning sacred animals to the entirety of the animal cults, and more specifically to the most famous sacred animal, the Apis.

77That there was some knowledge of hasye, and hence perhaps the deliberate production of hasye, is implied by a few other authors; as we have seen, both Herodotus and Aelian allude to the basic belief, and a third source is found in the North African writer Tertullian, who refers to different spirits that live in water, one of which is the esies, the spirit of “one whom water has drowned [lit. ‘killed’]”.²⁵²

78But perhaps the most famous reference to the idea in antiquity is found in the death of a human animal: Antinous, the Bithynian youth who was the lover of the emperor Hadrian, and is reported to have died, most likely by drowning in the Nile, in the autumn of the year 130.²⁵³ Following his death, he was acclaimed as a god by Hadrian, who founded a city in his honour, and apparently identified the appearance of a new asterism which marked his divinisation.²⁵⁴ The breadth of his cult is attested by the nearly one hundred surviving statues of his likeness, often depicted with the trappings of other deities such as Dionysos or Osiris, alongside inscriptions, metal coins, remains of temples and statue bases, and other sundry pieces of evidence.²⁵⁵

79One key detail of his story, however, is that almost every one of the principle accounts suggests or insists that he did not die by accident, but was deliberately killed by Hadrian in some kind of ritual.²⁵⁶ The purpose of the ritual is unclear; one of the clearest statements is in Aurelius Victor’s De Caesaribus, in which the author tells us that Hadrian killed Antinous “desiring to extend his own life”,²⁵⁷ and this seems to have been the implication of the author of the Historia Augusta, which describes Antinous as a devotum.²⁵⁸ This would draw upon a longstanding topos in Greek, and by extension Roman, culture, that it was noble for the subordinate partner of a relationship to sacrifice themselves for their husband or lover – the most famous example of this idea is found in Euripides’ Alcestis, in which the titular character willingly takes the place of her husband in order to spare him from his fated death.²⁵⁹ On this understanding, Hadrian, whom the same historians describe as excessively interested in religion and magic, may have received some kind of oracle which led him to fear for his life, and asked Antinous to die in his place.²⁶⁰

80A second interpretation is suggested by the writings of the Neoplatonist Celsus, and his Christian opponent Origen; Celsus compares Jesus, a man who continued to produce miracles after his crucifixion, to Antinous, likewise a divinised dead man who was responsible for oracular and healing miracles after his death.²⁶¹ Origen is of course incensed at this comparison, as Celsus would have intended – Origen notes with disgust that Antinous suffered from the “female malady”, and Origen is clear that someone who had gratified such desires had nothing in common with Jesus.²⁶² Instead, he says, the case of Antinous is the same as that in other temples of the Egyptians, where the priests, by their magical arts, set up demons in temples where they perform apparent miracles.²⁶³ The implication here is clear – he accuses Egyptian magician-priests of setting up a false oracular spirit, and if we take away the particularly Christian accusation that the spirit was a false demon, we can see an echo of this in a statement of the Historia Augusta that Hadrian claimed to receive oracles through Antinous.²⁶⁴

81One of the fullest sources for the cult of Antinous, the obelisk which is now in Rome, and which contains a hymn to the god written in hieroglyphic script in Traditional Egyptian, confirms that Antinous was believed to appear in dreams to his worshippers, and to heal the sick.²⁶⁵ In this he was not unique; many other gods of this period in Egypt and elsewhere were the subject of incubation cults in which individuals would sleep in their shrines in order to dream prophetic and healing dreams; notable examples include Asclepius, Serapis, and Isis. Gil Rendberg has persuasively argued that his cult may have begun with a prophetic dream – perhaps the source of the “oracles” mentioned in the Historia Augusta – in which the bereft Hadrian dreamt of his dead lover, an event which subsequently led to the establishment of the cult and city in his honour. ²⁶⁶

82If, as I have suggested, the concept of the magical hasye was circulating in Egypt in the early centuries of our era, it is easy to see how the death of a man in the Nile, followed by his deification, and his appearance in oracular dreams, might spark rumours that he had been deliberately killed and set up as an oracular spirit; as we have seen, there are surviving rituals attested on papyri only a century later in which animals were killed in order to send dreams, or to function as divine assistants.

83Many authors have taken the accounts of these ancient historians seriously; they are, after all, almost unanimous in their claim that the death was deliberate, and the date of the founding of Antinoopolis, 30 October, was six days after the festival which commemorated the death of Osiris, probably an auspicious day to produce a human hasye, with the discrepancy perhaps explained by the time it took for Hadrian to grieve and organise the foundation. Royston Lambert even dramatises Antinous’ internal struggle before his death, as he wonders how long it will be before he grows too old for Hadrian, what will happen if the emperor dies as an oracle has foretold, and if he will indeed be reborn as an Osiris.²⁶⁷ Ultimately, of course, it is unknowable if Antinous’ death was deliberate, although it seems more likely to me that it was accidental; the same authors who accuse Hadrian of his death disapproved, as we have seen, of his religious and sexual proclivities, and might easily add to these scandals that of human sacrifice.²⁶⁸

84But there was a clear difference between drowning a falcon or cat and drowning Antinous; while animal death was, in many respects, a routine part of life, the deliberate killing of a human, outside the context of warfare or execution, was shocking. And this difference seems to have had another, subtler consequence: the insistence in every account that the death of Antinous was voluntary.

85Why this insistence? The theme of a substitutory sacrifice to save a life was present in Graeco-Roman culture, but the voluntary variant existed alongside involuntary versions.²⁶⁹ If the primary inspiration was Egyptian, the death of Osiris was not voluntary; the episode required the punishment, even if only temporary, of his killer, Seth, and this is arguably the reason that the ritualist is almost always invisible in the killing rituals, to avoid taking on this role. The death of the animals in the magical papyri is not voluntary, although the question does not really arise; as we have seen, they are examples of “deading life” and their interests in life are secondary to their use in death. Unlike individual humans, and to a lesser extent, the animal gods such as the Apis, who have at least a “collective individuality” (each Apis bull is unique while it is alive), the animal hasye and votive mummies are generic. They are important not as individuals, but as a class; as the type “falcon” or “cat” they are collectively sacred to particular gods. But Antinous was not generic; his importance came not only from his specific relationship to Hadrian, but also from his personal history, up to and including his death. He would presumably have retained this same specificity even as a god, this individual identity which is inscribed, in an idealised form, in every one of his portraits. If he had been killed unwillingly, might he not have become one of the restless spirits who haunt Graeco-Roman legend? Might not his deified spirit seek revenge on his killer?²⁷⁰ The voluntary nature of his death might have been a necessary inference from the fact that this did not seem to happen: it might have been reasoned that if the dead and deified Antinous was well-disposed towards Hadrian, and this deified Antinous was a continuation of the living Antinous, then the living Antinous must have consented to his own death.

86But what of the animal hasye, who are given no individuality, and who did not consent to their deaths? Is there no vengeance for them? In a work well known in animal studies, Derrida discusses his cat, and, sensitive to the human tendency to genericise animals, insists,

I must immediately make it clear, the cat I am talking about is a real cat, truly, believe me, a little cat. It isn’t the figure of a cat. It doesn’t silently enter the bedroom as an allegory for all the cats on the earth, the felines that traverse our myths and religions, literature and fables. There are so many of them… No, no, my cat, the cat that looks at me in my bedroom or bathroom… does not appear here to represent, like an ambassador, the immense symbolic responsibility with which our culture has always charged the feline race… If I say ‘‘it is a real cat’’ that sees me naked, this is in order to mark its unsubstitutable singularity. When it responds in its name… it doesn’t do so as the exemplar of a species called “cat”, even less so of an “animal” genus or kingdom. It is true that I identify it as a male or female cat. But even before that identification, it comes to me as this irreplaceable living being that one day enters my space, into this place where it can encounter me, see me, even see me naked. Nothing can ever rob me of the certainty that what we have here is an existence that refuses to be conceptualized. And a mortal existence, for from the moment that it has a name, its name survives it.²⁷¹

87It is to this insistence on the particularity of one cat to which I will finally turn. It is true that the agency of the animal god is denied in almost every account, from the votive mummies of the animal cults to the magical hasye of the Roman-period texts, but in this “almost” there is a single exception.

88In the curse in which a cat is drowned, mummified, and then buried, the formula not only acknowledges the cat’s death as unwilling, but urges him to take vengeance.²⁷² The killing of the cat is described as his being “mistreated”, ²⁷³and his death is called an “outrage”.²⁷⁴ He is called upon to “rouse himself”,²⁷⁵ to “revenge himself upon [his enemies]”, ²⁷⁶to “destroy and ravage in the coming dawn”.²⁷⁷ This is a radical departure from the other texts, in which the death of the animal is present only as a mythic or ritual event, and not as a specific harm done to a specific victim. In part this acknowledgement is made possible by the spell’s purpose – it is a curse – but there also seems to be a degree of syncretism here between the Egyptian concept of the hasye and the Greek concept of the biaiothanatos, the violently-dead who might become restless, vengeful spirits. Nonetheless, the recognition of the cat as an individual, and of his death as a particular event, is incomplete; this is, of course, a recipe, and not the account of a ritual which was performed, and so it must be to some extent generic. But the cat is still seen primarily as the symbol and incarnation of the solar-typhonian god. More strikingly, there is a rhetorical sleight of hand here – the killer, the ritualist, remains invisible, and the “outrage” of the cat’s death is blamed on the curse’s victim.²⁷⁸ It is against this victim that the rage of the cat-faced god is directed, and not the hands which might have held his head under water until his flailing body stilled.

Conclusions

89The complex of ideas that gave rise to the concept of the hasye are both particular and universal – the history of the animal cults, the centrality of the Nile to life in Egypt and the danger of death in its waters, the knowledge of mortality and the hope of immortality. But within the specific configuration of these ideas we can see the ways in which distinctions were made between particular classes of being.

90Every being – god, human, and animal – was subject to death. For the sun, this death was cyclical – he was born each morning in the east, aged and died in the west, and traversed the underworld at night. Similarly, the stars of the ecliptic accompanied the sun, but underwent a more complex cycle in which they suffered seventy days of purification in the primeval waters before they were reborn. Osiris, and other osirian deities, died once, but having undergone funeral rites could continue to project their power from the underworld, serving as eternal ancestors to humans and gods.²⁷⁹ A human death was both linear and personal – a transitional, vulnerable moment in which they could, depending upon their moral character and ritual preparation, transition to a new state of existence, as an Osiris, a Hathor, an akh, or a hasye. This transfiguration might assimilate them to some extent with the greater divine powers, but it did not efface their identities, which remained intact through their names, the memory of their lives (which might be commemorated in inscriptions), and their relationships with other individuals, both living and dead.²⁸⁰

91But in Egyptian religion, the death of animals was almost always generic. Again, almost; in the phenomenon of pet mummies,²⁸¹ as well as the later shared graves of the human poor and their dogs,²⁸² we get some sense of individual animals, with their own names, histories and relationships. But for the most part the votive mummies, the animal hasye, and even the animal gods, are generic; they are divine symbols, and if their lives had pleasures and sufferings, or if their deaths were peaceful or horrific, these were rarely acknowledged or spoken of. It is when they died that they began to live, when they learned to listen to prayers and to send dreams.