Inscribed Lead Tablets in Small Animal Bodies: An Intersection of Greek and Egyptian Cursing in Late-Roman Egypt?

1 Animals in Greek and Egyptian Cursing

To fully understand this fairly complex combination of Greek and Egyptian rituals, we begin by stressing how in their cursing rituals the Greeks and Egyptians deployed small animals in different ways. The use of animals as effigies was fairly widespread in the Greek world, already in archaic times, when they were used in conditional self-curses attached to oaths.^[9] More than a millennium later, we find similar practices coordinated with curse tablets, for example, a lead curse tablet from a second-century CE grave in Carthage (DT 241.15 – 20) that begins by adjuring a number of superhuman entities to bind the limbs and tendons of a charioteer named Victoricus and all of his horses, and then adds: “Just as this rooster has been bound down in its feet, hands, and head, so, too, bind down the legs, hands, head, and heart of Victoricus the charioteer.”^[10] The use of the deictic pronominal adjective “this” shows that a trussed up rooster must have been placed in the grave near the tablet. Sometimes the manipulation of the animal focuses on a single part of its body, for example, in two defixiones inscribed in Latin that use the tongue of an animal as the focal point for silencing their victims. The first from Carthage reads “[Just] as I have ripped out the tongue of this rooster … and pierced it, so too may they (i. e., the ghosts of the people buried in the tomb) silence the tongues of my enemies!” The second is a recently published curse tablet from a town near Seville: “Just as a frog without its tongue is quiet and silent, so, too, may Marcellus be mute, silent and debilitated against Licinius Gallus!”^[11] There is also evidence that such effigies could be used without a written curse and placed above ground near the victim, rather than in a grave. In his autobiography, the fourth-century CE orator Libanius tells us that, after he had become sick with headaches and gout and was unable to read, write, or speak before his students, the mutilated and contorted body of a chameleon was discovered in his lecture room “with its head tucked between its hind legs, one of its front legs missing, and the other closing its mouth for silence” (Orat. 1.249).^[12] The unnatural contortion of the chameleon’s body seems to have been interpreted by Libanius or his friends as the cause of his gout, but the treatment of its mouth and front legs suggests that the curse had also been designed to attack his abilities as an orator.^[13]

The Egyptians, on the other hand, seem to have placed curses into the mouths of human mummies or dead animals, as we see in this Demotic recipe for an erotic curse dating to the second or third century CE (GEMF 16.1075):

It brings a woman also. You should write this name on the reed leaf with blood of a … or a hoopoe; you should put the hair of the woman in the leaf; you should put it (i. e., the leaf) in the mouth of the mummy; you should write this name on the ground saying, “Bring NN, daughter of NN, to the house, to the sleeping place in which NN, the son of NN, is!

Presumably the command inscribed on the ground is addressed to the ghost of the mummified person, into whose mouth the woman’s hair has been deposited, along with a leaf also inscribed with a powerful name. We see a similar practice at work two centuries later in a fourth- or fifth-century CE papyrus found in Hermopolis with brown hair attached to it (PGM XIXa): at the end of a long litany of magical names we find this command (15 – 16): “arouse yourself for me, nekudaimon, … and fulfill what has been inscribed and inserted into your mouth, immediately, immediately.”^[14] A curious phrase near the end of curse suggests, moreover, that the ghost is under compulsion as long as the papyrus remains in the mouth of his corpse.^[15]

Four other recipes from Roman Egypt replace the human mummy with a dead dog or cat, again primarily for erotic curses. The parallels are most obvious in a Greek recipe from the Fayum (PGM XXXVI 361 – 371), which has the rubric “Irresistible love-binding procedure (philtrokatadesmos).” It instructs us to inscribe the curse on donkey-skin, put the ousia from the victim (i. e., her fingernails or hair) inside of it, along with the “the plant that compels” (katanagkê),^[16] and then “place it in the mouth of a dead dog.” The curse itself asks a certain sisisôth to “attract her, NN, to me on this very day, in this very hour” and then adjures him by some more powerful entity with a long magical name.^[17] The two recipes discussed above suggest that he is the spirit of the dog, or perhaps even Anubis himself, but there is no way to be sure. A fragmentary Greek recipe of slightly earlier date probably contained similar instructions, but the text of the curse is lost.^[18] The dog, however, is described as “violently killed” (biaothanatos), which is typical of the human ghosts that are used for cursing, once again raising the strong possibility that the angry spirit of the dead dog would be called upon to do the cursing.

The earliest example of using the mouth of a dead dog as a receptacle for a curse is a Demotic recipe for sending dreams, a genre that in the later handbooks is primarily concerned with erotic submission (GEMF 17. 112 – 116):^[19]

On a new papyrus you should draw an image of Anubis with the blood of a black do[g] on it, you should write these writings under it, you should put it into the mouth of a black dog of the embalming house, you should make great offerings before it, you should put incense on the brazier before it, you should make a libation of milk of a black cow or a spirit whose face is grim(?)^[20] and you should put its recitation in its mouth.

The “recitation” (101 – 111) invokes Anubis to go into the underworld and allow a spirit to come into the dreams of the victim in the guise of a god that the victim reveres and to tell the victim what to do.^[21] Here, once again, instead of placing the message into the mouth of a human mummy, we are told to insert it into the mouth of “a black dog of the embalming house”. Given the recipe quoted above from the Fayyum formulary, this is likely the mouth of a mummified dog,^[22] but it may perhaps have been a statue of a black dog, which, like the image drawn on the papyrus, would be of Anubis himself. This second possibility fits better with the pattern that we saw in the curses placed in the mouths of the mummies, which were addressed to the mummy itself.^[23] A second-century CE bilingual handbook from Thebes contains a similar dream-sending recipe (this one in Greek) that uses a different animal (GEMF 15. 156 – 169 = PGM XII 107 – 121): “Take a completely black cat that died a violent death, fashion a strip of papyrus and inscribe on it with myrrh ink(?) the following things and the name of the person to whom you want to send the dream, and place it in the mouth of the cat.” The curse that follows invokes a series of magical names to send the dream, but their identity is not at all clear.^[24]

These two animals, the dog and the cat, were among the most popular animals mummified in Egypt, especially during the Roman period, when they were used as votives offered to the god that they represented or incarnated. The Egyptians, moreover, sometimes gave these mummies offerings and prayed to them for favors, a practice which seems to lie behind the instructions quoted above for placing requests in their mouths.^[25] One wonders, in fact, whether behind the choice of the mouth as repository lies a traditional “Opening the Mouth” ceremony performed by Egyptians on human and animal mummies and statues, down through the Roman period, a ritual that was designed in part to put breath into these objects and make them come alive and presumably be more attentive to mortal requests for help.^[26] It is important to note, moreover, that all of the texts discussed above that were inserted into the mouth of a dead human or animal were inscribed on media other than lead: a reed leaf, donkey skin, or papyrus. This is in great contrast, of course, to the Greek practice found throughout the Roman Empire of pairing a lead curse tablet with a dead or dying animal, which clearly serves as an effigy for the victim of the curse. As we will see, two curse recipes from Roman Egypt seem to combine these two traditions in a novel manner, because they involve inserting inscribed metal tablets of Greek design into small animals in elaborate rituals that reflect or distort the Egyptian traditions for creating and using animal mummies. Both recipes are preserved in fourth-century CE formularies and struggle in different ways to bring the two traditions together.

2 The Curse of the Desiccating Frog

The frog recipe from the Fayum is short and without any rubric or advertisement (PGM XXXVI 231 – 255):

Take a lead lamella and inscribe with a bronze stylus the names and the image given below. And after smearing it with blood from a bat, roll up the lamella in the usual fashion. Cut open a frog^[27] and put (the tablet) into its stomach. And after stitching it up with Anubian thread (and) a bronze needle, hang it up on a reed from your property by means of hairs from the tip of the tail of a black ox, at the east of the property near the risings of the sun.

The command to inscribe a lead tablet with a curse and then roll it up finds many parallels among the lead curse tablets that survive from Roman times, but the rest of it is alien to Greek practice, especially the placement of the curse tablet inside of the stomach of an animal and the deposition of the entire apparatus out in the open air, which is contrary to the Greek practice examined earlier of burying curse tablets in graves or depositing them in underground bodies of water.^[28]

Fig. 1: Model or diagram for how “the names and the image” are to be arranged on the tablet. PGM XXXVI 231 – 255.

After his instructions on how to treat the frog, the Oslo scribe provides us with a kind of model or diagram for how “the names and the image” are to be arranged on the tablet (Figure 1). The letters are smaller than those in the instructions above and the tablet itself is designed in landscape, rather than letter format, which is unusual among the extant curse tablets and also seems to be a late third- or fourth-century development.^[29] We see at the center of the design a linked pair of images, a larger one with the profile head of a bird, who holds the smaller, frontal figure by the top of the head. Although the scribe renders the three magical names poorly, it is clear that he wanted to position ousiri, then seseggen barpharanges, and then the erekisêphthê-palindrome across the top of the tablet and then repeat these words in each successive line, while reducing them letter by letter, so that ousiri would be transformed into a “wing formation” (a downward-pointing right triangle) by removing one letter from the beginning of the name and so that seseggen barpharanges and then the erekisêphthê-palindrome would be transformed into heart-shaped formats (downward-pointing isosceles triangles) by simultaneously removing one letter from the beginning and end of the name at the same time.^[30] To the left of the bird-headed figure, the scribe unsuccessfully split the heart-formation of seseggen barpharanges on either side of the upraised sword and eventually got confused. But on the right side, he did a much better job with the palindrome, from which he successfully removed the first and last letters for the first three iterations, below which we see centered the word bathron, which means “step” or “rung of a ladder”, although here it apparently serves as a scribal note that means something like “keep going step by step”.^[31]

Below this scribal note, we finally see the text of the curse itself:

Lord angels, just as this frog drips with blood and dries up, so also will the body of him, NN, whom NN bore, because I adjure you, by those who have been put in command of fire, maskelli maskellô (add the other usual items).^[32]

Although in the handbook diagram (Figure 1), the text of this curse fits nicely within the rectangular format of the tablet, this was not the intent of the author: the instructions say to inscribe only “the names and the image” on the tablet and, if the instructions for the heart-formation of the erekisêphthê-palindrome are followed downwards for another dozen or so lines, there would be no room for the curse in the lower right corner of the tablet. This suggests that the curse itself was either spoken out loud or inscribed on the back of the tablet. The text of this curse reflects the simple and straightforward kind of effigy magic that we saw in in the previous section, in which the formula equates the desiccating body of the frog with that of the victim. There are, however, two important differences: here the tablet is inserted into the body of the effigy, rather than placed beside it, and both are then deposited together in the open, before the rising sun, rather than in a grave or an underground body of water.

Stranger still is the instruction to make an incision in the frog’s body and then repair it with “Anubian thread” (λίνῳ ἀνουβιακῷ), details that might, in fact, allude to mummification, because Anubis was the god who presided over mummification^[33] and because the very first step in the process of mummification was to make a vertical incision on the left side of the abdomen. Although Herodotus, in his otherwise fairly accurate description of Egyptian mummification, says that this wound was eventually sewn up, there is little evidence, in fact, that this is so.^[34] And generally speaking, small animals were not cut open and eviscerated, with the exception sometimes of dogs and cats.^[35] And from this Egyptian perspective, the choice of animal is also a bit odd for a curse. In Egypt, the frog was traditionally viewed and treated as a fertility symbol—closely connected with the Nile flood, which was annually followed by the birth of huge numbers of frogs from the recently deposited alluvial waters, and with the worship of Hequet, a goddess of rebirth and childbirth, who was depicted with the head of the frog.^[36] The scribe seems, perhaps, to have borrowed the frog used in Greco-Roman curses as an effigy and to have deployed it in a traditional Egyptian ritual, replacing the usual dog- or cat-mummy. But by placing it in the hot sun at the edge of cultivated land, and not the fertile Nile mud, he reversed or perhaps even perverted the frog’s usually positive associations in Egyptian thought.

The presence of the rising sun as a kind of agent for the curse also may reflect a tradition of Egyptian cursing, one that uses the unrelenting heat of the sun as a key part of the curse, much like the Greeks used burning incense or lamps or the hottest parts of the Roman bath.^[37] The best evidence such curses comes from another recipe in this same Oslo handbook, PGM XXXVI 102 – 133:

Take clean papyrus and with myrrh ink write the names and image given below and say the spell three times. The names and the figure to be written are these: “Hear me, you, who build and destroy and who became the mighty god, whom a white sow bore, althaka eiathallatha salaioth, who appeared in Pelousion, in Heliopolis possessing an iron staff with which you walled up the sea (i. e., on both sides) and passed though,^[38] … (you), who completely dried up all the plants. Attract to me, NN, her, NN, aflame, on fire, flying through the air, hungry, thirsty, not finding sleep, loving me, NN whom NN bore, until she comes and glues her female pudenda to my male one, immediately, immediately; quickly, quickly”.

Aside from the mysterious reference to the white sow,^[39] the god invoked here seems to be the Egyptian sun-god worshipped in Pelousion as Amun-Re and in Heliopolis as Re,^[40] both places in Lower Egypt and presumably well known to the compiler of the Oslo papyrus, which was acquired in the Fayum and in all likelihood came from there. Amun-Re may also be equated here, as often happens in Roman-period magical texts, with Yahweh, whom in this case the scribe links to an Egyptian context by referring to the parting of the Red Sea.^[41] The Oslo scribe did not include any instructions for this recipe, but the rubric (“another burning procedure”) tells us that this recipe is another version of the previous one (ll. 69 – 101), in which we are told to glue the inscribed papyrus to the dry vaulted ceiling of the hot room in a Roman bath. This seems, therefore, to be another case of an Egyptian curse asking the sun to set a woman on fire with lust, but it uses a similia similibus mechanism triggered not by the Egyptian sun, but rather by the hottest part of a Roman bath.

Fig. 2: Papyrus inscribed with an elaborate design in landscape format. PGM XXXVI 102 – 133.

The prayer to the Egyptian sun-gods, when spoken aloud, would have sufficed all by itself to curse the victim, but this recipe (again like the frog recipe) adds a written component: a papyrus inscribed with an elaborate design in landscape format (Figure 2) with a central figure, who again holds his victim by the hair and who in this case is surrounded by four magical words, all of them in wing-formation. Then, to the far right, we see a shortened form of the same prayer: “Attract to me, NN, her, NN, aflame, on fire, flying through the air, loving me, NN whom NN bore, immediately, immediately; quickly, quickly! Accomplish it!” This abridged version of the prayer has, however, been completely stripped of the local Egyptian names and details and we must assume that it is addressed to the central figure in the design and the magical names that surround him. In both of these Oslo recipes, then, the scribe takes a Greek curse inscribed on lead or papyrus and frames it in a new, Egyptian setting, that involves the sun-god, albeit in different ways, as a key element or agent in the curse.

Fig. 3: Detail of PGM XXXVI 231 – 255.

Fig. 4: Drawing of a frog from Knobel (1896) 39.

There are, finally, some signs that in producing this composite recipe for the frog curse, the author may have tried to make the central scene on the lead tablet conform to the framing ritual into which it was inserted. Although the rendition of the frontal face of the victim on the tablet is quite similar to that of the victim depicted on the papyrus used in the fire spell (Figure 2), we can see that they differ in the treatment of the hands and legs. In the enlargement of the Oslo drawing (Figure 3), we can see that the “victim” has strange spider-like fingers that are tipped slightly to our left in a starburst configuration. I suggest that in producing this drawing the author did not aim to represent human hands, but rather he made a rudimentary attempt to depict the front feet of a frog (Figure 4), and that he drew circular rings around the rear legs that roughly mimic the rings one finds on the rear legs of frogs. It is perhaps possible, therefore, that with regard to the image of the victim drawn on the lead tablet, the scribe aimed at depicting a human-headed frog in a scene that roughly depicts the ritual described earlier in the recipe: a dominant bird-headed figure (the practitioner) holds in his left hand the head of a frog-bodied man (the victim), an action that exposes his belly. And he holds in his right hand the sword with which he can cut it open.

3 The Curse of the Cat-Mummy

We see an even more telling example of this intimate relationship between curse tablet and small animal in an elaborate Greek recipe preserved in a somewhat damaged papyrus roll in Paris (GEMF 55. 131 – 300 / PGM III 1 – 164). The procedure is framed by a traditional Egyptian recipe for drowning a cat, wrapping it like a mummy, and then asking a “cat-faced daimôn”^[42] (the spirit of the deified-cat, who at times seems closely connected to Helios)^[43] to “do the NN thing”. It is, in short, an “all-purpose” recipe that can be used to obtain many different goals. But this recipe also contains instructions for inscribing four objects to be used in the procedure: three curse tablets (probably of lead) to be inserted into the dead cat’s anus, larynx, and one other orifice (lost in a lacuna) and a papyrus with additional curses, that is to be wrapped around the cat. As in the case of the frog curse, the author seems to have created this composite recipe by combining two different cultural practices: (i) a traditional Egyptian ritual designed to create a cat-mummy, which could be asked to do whatever the practitioner wished (i. e., “to do the NN thing”); and (ii) series of traditional Greek curses, some of which aim at binding rival charioteers. This somewhat awkward combination becomes clear if we note how the recipe begins and ends:^[44]

[Take] a cat and make an esiēs^[45] by throwing (it) [into] the water. [While] you are drowning (it), speak to [its] back. The formula over the drowning: “Hither to me, (you who) control the highest form of Helios, (you) the [cat]-faced god… (GEMF 55. 131 – 133 / PGM III 1 – 34).

This is the procedure of the cat, concerning [every] purpose: a restraining procedure for charioteers in a competition and a dream-sender and a philtrokatadesmon, a separation procedure and one for causing enmity. (GEMF 55. 297 – 300 / PGM III 1 – 34).

The recipe thus begins, like the frog recipe, with no initial rubric, but rather with a series of instructions. Only at the very end do we find the title of the recipe in the colophon, which also tells us the purpose of the procedure^[46] As it stands, this colophon is contradictory or redundant, because it initially claims that the procedure is good for “every action”, but then adds the four specific goals that we see listed above, most of which were popular goals on extant Roman-era curse tablets or in the recipes for them. Of these individual goals, it is significant, as we will see, that the first mentioned is to restrain charioteers.

We can see the contours of the original Egyptian drowning ritual most clearly, when we compare this GEMF 55 / PGM III procedure with an earlier recipe in GEMF 31 / PGM I 1 – 42 that involves the use of a mummified falcon:

From the comparison above, we can see how the main elements of the cat-recipe correspond to the Egyptian ritual whereby an animal is drowned, wrapped as a mummy, set up in tomb or mortuary temple, and offered a form of sacrifice, a sequence that we also saw in GEMF 17. 112 – 116, the most Egyptian of the recipes discussed in Section 1, which directs us to inscribe a papyrus with a request, to put it in the mouth of a “black dog of the embalming house,” and then to make offerings of incense and milk. And as the parallel ritual with the falcon suggests, in the original Egyptian recipe the cat is “deified” by the rite and becomes itself the object of veneration, which is given offerings and which, as the cat-faced daimôn, becomes the recipient of oral requests to which it can respond.^[47]

These recipes for the cat and falcon mummies differ, however, in one important way: the dead cat is stuffed with and wrapped in Greek curses that aim at charioteers (GEMF 145 – 156 / PGM III 40 – 43):

Take the (i. e., drowned) cat and make [three] tablets, (placing) one in its anus, one in its [mouth],^[48] and one in its larynx. And [concerning] the deed (i. e., that you want performed), write the formula with vermilion ink on a new papyrus, concerning^[49] the chariots, the drivers, the chariot-boards,^[50] and the lone-running horses.^[51] Wrap (the papyrus) around the body of the cat and bury (it). Light 7 lamps upon 7 unbaked bricks and make sacrifice to it (i. e., the cat-mummy) by burning storax and be of good cheer. Keep its body and [carefully guard?] it, either in a tomb or in … the place of ….

Here we find the first explicit mention of chariots and charioteers, and we should note that the four other goals mentioned in the colophon (dreams, binding, separation, and hatred) are completely absent. Unfortunately, the verbatim “formula” to be inscribed on the papyrus wrapping seems to have gone missing, but it must have directed the “cat-faced daimôn” to attack rival charioteers in a forthcoming race, like the scores of lead tablets that have been unearthed in Carthage, Rome, and beyond.^[52]

Chariot racing is raised again in the description of the ritual sprinkling of the water used in the drowning and the oral invocation that accompanies it (GEMF 170 – 183 / PGM III 40 – 43):

Take up the water from the drowning, sprinkle it [on] the racecourse or in the place where you are performing (the procedure). [This] is the formula spoken to the water from the drowning, while you are sprinkling (it): “I call upon you, genetrix of all mortals, you who pressed together in your arms the (i. e., scattered) limbs of Meliouchos and Meliouchos himself, orobastria neboutosoualêth arkuia nekuia ermêekatêlthê ermekatêlêth amoum amoutermuôr. I adjure you, daimōn, who in this very place is being made perfect, and (I adjure) you the daimōn of the cat. Come to me on this day today and from this very hour and perform for me the NN deed. (Add the) usual, whatever you wish.

The text in the second half of the passage contains two disconnected speech-acts, which invoke a composite lunar goddess, who (i. e., as Isis) once gathered and reassembled the limbs of Meliouchos (i. e., Osiris), and who, as is often the case in the Roman period, is equated with moon goddesses such as Artemis, Selene, and Hekate. The request to this goddess seems to have gone missing but given that the prayer focuses on how Isis once brought Osiris back to life, perhaps she was asked to revive the cat-mummy (who is adjured in the next sentence), so it can perform the curse. The incomplete adjuration^[53] at the end of the passage is, however, directed at two different daimones, the one who is “in this place” and another “of the cat “, who are then individually adjured to “perform (182: ποίηϲό̣[ν) for me the NN deed.” The singular form of this final imperative addressed to both daimones suggests, however, some textual confusion at this point. In the Egyptian ritual, the practitioner, by sprinkling the water at the racetrack, can presumably release this spirit (pneuma) into the hippodrome,^[54] where it can attack the charioteers who are being targeted by the curse. This “cat-faced daimôn” would have been sufficient for the job, of course, but the author of this recipe felt the need to add a second agent typical of Greek curse tablets, especially those of the Roman period, which often address the nekudaimôn or ghost, who hangs about a grave, although here the addressee seems to be a ghost is “in this place” (i. e., haunting the stadium), presumably a ghost of a charioteer who died before his time in some violent accident and who remains unhappily in the area.^[55]

Fig. 5: Drawing and text of Recto Column iii of PGM III by Raquel Martín Hernández.

The recipe then tells us what to inscribe on two of the three tablets inserted into the cat, but the instructions for the third text were apparently never copied into the recipe. The diagram for the second tablet—it is more fully preserved—combines, like the one in the frog recipe, names, images, and a curse (Figure 5).^[56] First, the scribe wrote in the center of the top line: “[On the 2^nd] tablet [this] is placed …”, followed by a version of a well-known palindrome that can be fully reconstructed from the many parallels,^[57] a[beram]enthôu[uth]leraex[an]-axe[thr]elthuoôethne[mareba]. He then added below two different instructions, each on its own line: “(Inscribe it) like a heart, like a grape cluster,” instructions which, like the word bathron in the frog recipe, remind us to write out the palindrome numerous times, taking care to remove the first and last letter from each side until only single letter—in this case a nu—remains.^[58] To make sure that the reader understood these somewhat technical terms, the scribe added below them a diagonal line that goes down to the right and the then up again, tracing the outline of a downward-pointing isosceles triangle that will be produced, if the diminishing name is correctly written out. On either side of this triangular guideline, we find a pair of indented titles or labels (the second reconstructed from the first): “right skeletos” to the left and “[le]ft [skeleto]s” to the right.^[59] And below each of these labels, we find an image of a naked man. The one on the left has an oversized crown(?) hovering high above his right shoulder and he is rendered slightly larger than the figure on the right; both hold whips aloft in their right hands and seem to have crossed bands over their chests like charioteers. The word skeletos in Greek originally referred to a “withered” or “mummified body” but it came to refer also to a “skeleton” in the modern sense, although the figures drawn on the papyrus do not seem to fit either definition. They are probably effigies for the charioteers, who are referred to in the plural as “my enemies” in the curse beneath the drawing on the right.^[60]

Further to the right of the righthand “skeleton”, we see an even larger, profile image of an animal-headed figure. There has been a long debate over the species of the animal head on this larger figure, some claiming it is of an ass or horse,^[61] but the repeated invocations in this recipe to the “cat-faced daimôn” suggests that the person who made the drawing intended it to have the face of a cat. Like the two “skeletons”, the cat-faced figure holds a charioteer’s whip aloft in its right hand, while gripping under his left armpit a long staff. The whip in the hand of this larger figure suggests that he is the ghost of the charioteer “in this place”, who is assimilated with the cat-faced daimôn as the agent of the curse.^[62] Below these three figures, there is text in two columns, which was presumably to be inscribed on the tablet: to the left in a narrow column the well-known iô-erbêth-logos ending in the request “perform the NN deed”; and to the right in the wider column beneath the cat-faced person, a long adjuration that begins:

I adjure you, the one in this place, the powerful and mighty messenger of this animal: rouse yourself and perform the NN deed, both on this very day and in every hour and day; rouse yourself for me against my enemies, NN, and perform the NN deed (add the usual).

Both deictic modifiers (“the one in this place” and “the powerful messenger of this animal”) refer to the same individual and they recall the double adjuration of the superhuman agent during the drowning (quoted above) as “you, the daimôn, who is being roused up in this place, and you, the daimôn of the cat.” The elaborate curse tablet depicted in the diagram in Figure 5 could have been used by itself, of course, to achieve the goals of the practitioner by invoking the ghost in the stadium alone, but its efficacy has been doubled by inserting it into the mummy of a cat.

Let us return, finally, to the drawings that are to be etched onto the curse tablet. The invocation of the cat-faced daimôn as one of the twin agents in this curse suggests that this daimôn is represented by the larger image, who towers over the two skeletoi, who, in turn, are probably, as noted above, the effigies of the charioteer victims of the curse. Although the person who drew the larger figure did an excellent job on the clothing and the rest of the body, he fared rather poorly with the head. Indeed, a number of scholars have suggested that the figure has the head of a donkey, despite the short ears, a designation that they see connected to the invocation of the Sethian logos in the left margin below the drawing.^[63] Others have argued that the short ears point to a horse’s head, like the horse-headed demon on the contemporary charioteer curses from the Via Appia in Rome.^[64] Given the somewhat abraded and smudged state of the head, certainty is beyond our reach, but following the lead of Raquel Martín Hernández, I suggest that in his rendering of the larger figure, the creator of this diagram began to draw the larger, elongated horse- or donkey-head that he found in his model for the lead tablet, but then he decided at the last moment to adapt the face, at least, to align it closer with the repeated invocation of the “cat-faced demon.”^[65] This adaptation is quite similar, I suggest, to the one made to the image of the victim in the frog recipe, which as we saw in the previous section aimed at conforming the frontal-faced victim drawn on that tablet with the frog that was killed and desiccated during the ritual.

Conclusion

We have seen that both the short curse recipe employing the frog and the much longer one with the cat-mummy reveal yet another means by which the traditional Greek texts and designs for curse tablets might be introduced into an already free-standing curse of a different origin. In the Oslo recipe, the author perhaps began with a Greek recipe that combined in parallel two modes of cursing that we saw above in Section 1: an inscribed lead tablet and a frog effigy. But then he combined these two objects more intimately by sewing the lead tablet into the animal’s stomach with Anubian thread and leaving it in the hot Egyptian sun, actions that seem to reflect, albeit at some distance, the treatment of animal mummies in traditional Egyptian practice. This connection with mummification is much more obvious, as we saw, in the cat recipe, in which the author began with an all-purpose Egyptian recipe for the cat-mummy and then adapted it to the specific goal of binding charioteers, by inserting curse tablets into its body and wrapping it in a papyrus, all of which were inscribed with Greek texts and designs. In both cases, however, one wonders whether the violent treatment of the animal would have seemed to be a reversal or violation of traditional Egyptian ideas or rituals. The frog, a positive symbol of life and fertility, was cut open, sewn back up, and then hung in the sun to desiccate, far from the Nile mud, with some gestures toward mummification. In the case of the dead cat, rolled-up lead tablets were shoved down its throat and up its anus, a combination which recalls the contortion of Libanius’ chameleon, with its face twisted up between his own legs, but which may have seemed antithetical to the Egyptian reverence for small animal mummies as powerful votive objects offered to the gods they represent or incarnate.^[66] But as always, it is difficult to interpret individual intentionality in instructional texts like these that have been expanded and adapted over time. There were signs, finally, that in creating both the frog and cat recipes, the authors sought to lessen the dissonance between the Egyptian and Greek traditions, in one case by altering the hands and legs of the victim’s effigy on the tablet in order to reflect the use of the dying frog as an effigy in the ritual, and in the other case by altering the face of the dominant figure to reflect the coagency of the cat-faced daimôn with the ghost in the stadium.

Abbreviations

1. A. Audollent, Tabellae Defixionum (Paris, 1904).

2. C.A. Faraone and S. Torallas Tovar (eds.), The Greco-Egyptian Magical Formularies, vol. 1 (Berkeley, 2022).

3. H. D. Betz, The Greek Magical Papyri in Translation, Including the Demotic Spells, (Chicago. 1986; 1996³).

4. K. Preisendanz [and A. Henrichs], Papyri Graecae Magicae: Die Griechischen Zauberpapyri², 2 vols. (Stuttgart, 1973 – 1974).