Hostname: page-component-74656f9884-8r8nv Total loading time: 0 Render date: 2026-11-06T14:47:50.921Z Has data issue: false hasContentIssue false
  • English
  • Français

Roman Magic Figurines from the Western Provinces of the Roman Empire: An Archaeological Survey

Published online by Cambridge University Press:  08 April 2015

Magali Bailliot
Magali Bailliot*
Affiliation:
CNRS (umr 7041), Parismagallib@hotmail.com
Rights & Permissions [Opens in a new window]

Abstract

This paper deals with magic figurines from the Western provinces of the Roman Empire based on an inventory of twelve figurines and their archaeological context. It underlines the place of the figurines in the ritual of defixio and demonstrates that complex curse rituals such as those described in the Greek Magical Papyri (GMP) were not performed only in the Mediterranean basin. It also notes that these magic Western figurines are often found in important places (such as cities and large villas) and in late contexts.

Keywords

figurinesmagicdefixiocurse ritualsRoman Western provincesFishbourneGreek Magical Papyri

Information

Type
Articles
Information
Britannia , Volume 46 , November 2015 , pp. 93 - 110
Copyright
Copyright © The Author(s) 2015. Published by The Society for the Promotion of Roman Studies 

INTRODUCTION

Graeco-Roman magic is well known through literatureFootnote 1 and through such evidence as defixiones, magic figurines (‘voodoo dolls’), amulets and apotropaic representations of the evil eye depicted on mosaics or bas-reliefs.Footnote 2Defixiones, commonly called curse tablets and binding spells, are one of the most important sources. They are inscribed, small, thin sheets (generally of durable materials such as lead),Footnote 3 rolled up and sometimes pierced by one or more nailsFootnote 4 or placed with a ‘magic figurine’.

Defixiones were a form of magic practised all over the Graeco-Roman world that bound a victim to the desires of the practitioner (defigens). The ritual was private and performed to ensure victory in a legal dispute or competition (races, games), to gain another's affections, or to separate lovers.Footnote 5 Curse tablets were often placed in a grave with the soul of a dead person in order to accomplish an evil deed or in a locus which enabled contact with chthonic deities.Footnote 6 They have been traced back archaeologically to the six to fifth centuries b.c. in Greece and SicilyFootnote 7 and continued to be used until the sixth to seventh centuries a.d.Footnote 8 Despite such longevity, the premise of the binding curse remained unchanged.

It must be emphasised at the outset that this paper rejects conventional definitions of magic and aims to avoid any of the resulting misconceptions about ‘magic’.Footnote 9 In fact owing to the conflationFootnote 10 of ‘magic’ and other religious procedures such as devotio, for example,Footnote 11 it follows the thinking of Hendrik Versnel who favours a more flexible approach with definitions being adapted to each precise case.Footnote 12

It is well-known that in Rome too much interest in the supernatural was regarded with suspicion or was a pretext for a death sentence, according to the social and political context.Footnote 13 Nevertheless magic cannot be considered as a set of beliefs and rituals whose rules conflicted with the conventional rules of religion.Footnote 14 In the specific case of curses, magic used the ideology of religion and its practices such as sacrifices. While magic had different intentions from religion and implied different relations between human beings and gods with regard to the action of constraint,Footnote 15 it can still be argued that magic rituals corresponded to religious strategies which placed the victims in the hands of divine wrath because of sacrilegium and impurity.Footnote 16 Some curses called ‘judicial prayers’ indicate that the stolen goods (a priori robbed by the victims) belong to the gods, while other defixiones give the victims the status of a dead person. In such circumstances magical action was justified, because the stated sacrilegium and impurity forced the gods to punish the victim. In this way, magicians can be considered as ‘mediums’ who reported the crime to the gods and who were constrained to create the necessary curse in order to purify the community. A defixio expresses the idea in the words: ‘… except for (?) the one who wrote this curse and who destroyed the adversary, because he did not act of his will, but he was forced to do so by thieves.’Footnote 17

Considering all the weaknesses inherent in conventional definitions, this paper also believes that it is more productive to concentrate on the tangible evidence, which reflects the place of magic in the lives of the ancients, and to focus more closely on recent archaeological discoveries, which provide data permitting us to observe the rationality of magic.Footnote 18 Of course this point of view is based on the archaeological nature of the data, but it also takes into account the Greek Magical Papyri Footnote 19 which have provided significant information on the curse ritual (sequences, gestures, invocations and hymns)Footnote 20 as well as demonstrating — before the more recent archaeological discoveries — that the deposition of curse tablets required some sacrifices and occasional figurines depicting the targeted victims. This paper will also follow the analyses of Fritz Graf who notably disproved the assumption that transfixing figurines with needles was meant to hurt and who argued that such objects were not identical to the victims.Footnote 21

Generally magic figurines — fashioned from wax, lead and clay — are a few centimetres tall. Their limbs are often bound or twisted and can be pierced by nails or needles.Footnote 22 While such figurines are commonly seen as being specific to the Mediterranean basin,Footnote 23 recent discoveries show that they were also used in the Western provinces of the Roman Empire. Germany, Austria, Britain and France have yielded twelve figurines. Comparisons with identified Mediterranean figurines and descriptions given by the Greek Magical Papyri show that most are linked to magic even though they initially seem to be little more than ‘exotic’ surprises for archaeologists excavating domestic settlements. This short survey focused on the Western provinces sets out to complement the fundamental study and inventory of Greek, Etruscan and Roman ‘voodoo dolls’ by Christopher Faraone published in Reference Faraone and Obbink1991Footnote 24 and already supplemented by further discoveries.Footnote 25

ANTHROPOMORPHIC FIGURINES IN RITUALS

Beyond the dramatic portrayal of pierced dolls and violent formulae inscribed on magic sheets and literature,Footnote 26 the defixio ritual was primarily technical and performative.Footnote 27 This ritual also described as ‘persuasively analogical’ was based on a belief in the power of formulae and was designed to restrain and inhibit the victims.Footnote 28 Spells were deployed to encourage future action and were linked to the gestures or wished-for state of the victims. The aim of the ritual was to bind or nail them down (kαταδε~ιν in Greek and defigere in Latin). According to this logic, the nailing and twisting of figurines was supposed to render victims unable to resist. Victims were to be weakened in the same way as the figurines were tortured. The gestures represented on the figurines depended on the aims of the curse, so that twisting, binding or nailing could also be replaced by other treatments such as decapitationFootnote 29 and burning. In many amatory spells torture with fire was supposed to cause the victims to burn with passion.Footnote 30 A spell from Hadrumetum (Sousse/Tunisia) illustrates the case: ‘Let Vettia who begat Optata burn … with love and desire for me.’Footnote 31 The choice of the material employed to fashion figurines was in itself linked to the logic of this analogy. Clay, wax and lead are easily malleable. The properties of each can be connected with the wish to dominate the victims.

Figurines could also be replaced by animals, the choice being related to their qualities, especially weakness.Footnote 32 A defixio found in a grave excavated at Chagnon (France) and designed to silence legal adversaries is based on the defencelessness of a young animal removed from its mother and on the nailing that it presumably suffered.Footnote 33 It is also based on the analogy between the silence of the dead person from the tomb in which the curse was deposited and the incapacity of the victims to testify at court: ‘Just as the mother of this kitten is unable to defend it, so too may the legal advocates of these men be unable to defend them. Just as this kitten is unable to get up, so too may they be unable (to get up). May they be transfixed, just as this one is.’Footnote 34 Another example also expresses the process, making a parallel between the binding of a rooster and the fate of the targeted victim: ‘Just as this rooster has been bound by its feet, hands and head, so bind the legs and hands and head and heart of Victoricus, the charioteer of the Blue team, for tomorrow.’Footnote 35

A parallel can be drawn between the figurines and depictions of victims which appear on certain inscribed sheets. The tied legs and hands of figurinesFootnote 36 are reminiscent of some images of victims engraved on curse tablets, such as the example of a defixio from Carthage.Footnote 37 These drawings engraved next to the spells had the same function as bound, twisted and nailed figurinesFootnote 38 and were used to enhance the magic spells.Footnote 39 Moreover, figurines were sometimes placed in lead boxes, possibly representing coffins, as in the case of a small rectangular lead container from Athens kept in the Royal Museums of Art and History in Brussels and first published by Cumont.Footnote 40 Other examples include six third- to fourth-century wax dolls each enclosed in a single lead canister from the temple of Anna Perenna at the Piazza Euclide in Rome.Footnote 41 Piranomonte argued that these containers served to ‘isolate the victims, to undermine their social and moral integrity, and to express the fact that they have been “caught” by magic attack and are never to be free’.Footnote 42

A funerary meaning has also been suggested. It can be said that cursing was based on the symbolic separation of the living and the deceased performed during the funerary ritualsFootnote 43 and that its goal was a kind of anathêma (temporary or permanent) made in order to suspend the membership of the victims from the community of the living.Footnote 44 As an example one might note a sentence from a defixio found in a grave at Minturno (Italy), which seems to create a parallel between the fate of the victim and the religious and legal status of the dead person buried in the tomb: ‘As the dead man is received neither by gods nor humans, so may Rhodine be received by Marcus Licinius and have as much strength as the dead man who is buried here.’Footnote 45 The funerary and ‘separating’ character of the ritual can be perceived through other evidence, such as the parallels drawn between the aims of the written curses and actions carried out directly on the figurines and on the tablets (which contained the names of the victims): burying and hiding defixiones and figurines was presumably designed to ‘send a victim to hell’ far away from the community of the living. The use of lead could also have had a funerary meaning and an irrevocable character as shown in a defixio from Kreuznach (Crucinatum) which says: ‘As this lead (tablet) remains, Valetis Sinto is sent to hell forever.’Footnote 46

Nailing (in the sense of defigere and defixio) could have been in line with the ‘sealing’ and with the conclusion of the ritual. This hypothesis is supported by the Latin word clavus (key or nail),Footnote 47 itself related to clavis (key), which comes from the verb claudo (shut, enclose). Clavus is also linked to clausula which means ‘end, conclusion of something’.Footnote 48 Both hypotheses (isolation of the victims and exclusion from the community of the living) are ultimately compatible.Footnote 49

INVENTORY OF FIGURINES

Throughout this overview of the use of figurines in magic ritual, several distinguishing features emerge, such as nailing, burning, binding/twisting and the placing of objects in containers. They enable the identification of anthropomorphic figurines recently found in the Roman Western provinces and thus increase the corpus of magic figurines. Some figurines remain difficult to interpret because of their appearance or their state of preservation and sometimes because of the lack of information about their archaeological context. The inventory follows a geographical order: (1) Germany – (a) Upper Germany; (b) Raetia; (2) Austria (Upper Panonnia); (3) Britain (Britannia); (4) France – (a) Gallia Belgica; (b) Gallia Lugdunensis. Figurines are numbered F1 to F12.

(1) GERMANY

(a) Upper Germany (F1–F3)

Mainz (Rhineland-Palatinate)/Mogontiacum/sanctuary

Three second-century dolls shaped from clay were recovered from the sanctuary of Isis and Mater Magna in the centre of ancient Mogontiacum. They were published by Witteyer.Footnote 50

F1 (fig. 1a). The first figurine (10 cm tall) was roughly made of clay then partially baked. It has neither neck nor shoulders. The right leg and forearms are missing. The face is crude, and the eye cavities and nose seem to have been shaped using finger pressure. Despite the simple technique and the poor state of preservation, the genitals can be seen and indicate a male. The left side of the body shows holes which reveal that the doll has been pierced by needles.Footnote 51 The figurine comes from a well located outside the temenos which had probably been abandoned during the second century.Footnote 52 The object was probably intentionally (and partially) burnt. A parallel can be drawn with a figurine from Karanis (Egypt) presumably placed over fire in order to render the victim of the ritual inflamed with passion in simila similibus fashion.Footnote 53

FIG. 1.

(a) F1. Figurine from Mainz (after Witteyer Reference Witteyer2005, 111, fig. 4); (b) F3. Figurine from Mainz (Photo: Generaldirektion Kulturelles Erbe Rheinland-Pfalz, Landesarchäologie Mainz); (c) F4. Figurine from Eining (after Spindler Reference Spindler1984, 113, fig. 96 right); (d) F5. Figurine from Eining (after Spindler Reference Spindler1984, 113, fig. 96 left); (e) F6. Figurine from Straubing (after Spindler Reference Spindler1984, 114, fig. 98); (f) F7. Figurine from Straubing (after Spindler Reference Spindler1984, 114, fig. 97); (g) F8. Figurine from Petronell/Bad Deutsch-Altenburg (after Gassner Reference Gassner2010, 224, fig. 3, Gassner et al. Reference Gassner, Steigberger and Tober2011, pl. LXIX, fig. 18); (h) F9. Figurine from Fishbourne (after Bailliot and Symmons Reference Bailliot and Symmons2012; photo: M. Bailliot); (i) F10. Figurine from Reims (after Bailliot Reference Bailliot2010, 110, fig. 25e; drawing: M. Brunet (INRAP)); (j) F10. Figurine from Reims and box made with cups (after Bailliot Reference Bailliot2010, 97, 103, 105, fig. 23; photo: INRAP); (k) F12. Figurine from Piriac-sur-Mer (after Hervé-Monteil Reference Hervé-Monteil2010, 52).

F2. The second clay doll has disintegrated.Footnote 54 It was found in the fill of the same well as F1. The position tends to suggest they had been carefully placed.Footnote 55

F3 (fig. 1b). The third clay figurine (15 cm tall) is a representation of a victim mentioned on a nearby rolled-up inscribed sheet.Footnote 56 It is broken in two pieces. Its body is oval and its legs are very short. The forearms are missing. The head and neck are well formed but it has no shoulders. The eyes are depicted by two small holes, while the mouth is also represented by a simple notch. The nose and eye cavities have been shaped together by finger pressure. Its body has been pierced all over with needles.Footnote 57 Moreover it has both breasts and prominent male genitals. The dual gender of the figurine presumably had a ritual meaning. Romans used to interpret human beings with doubtful gender as a sign of divine wrath or evil that had to be removed by being cast out beyond the limits of the city. When such a child was born it was regarded as a prodigium Footnote 58 and was sentenced to death as a necessary purification of the city and of the community.Footnote 59 Latin literature is rich in references to hermaphroditism as a source of fear and as prodigia.Footnote 60 The phenomenon appears, for instance, sixteen times in Livy's The History of Rome and six times in Ovid's Metamorphoses (for men and women); Pliny the Elder also documented some sex changes. A similar ideological attitude can also be directly observed towards hermaphroditism, death and signs of evil. Such events were considered as impurities to which the response was a ritualised separation from the community of the living. The hermaphroditic character of the figurine may have been based on such beliefs in order to exclude the victim from the community of the living and to allow it to be cursed.

This figurine comes from a ditch previously filled with rubbish. It was found next to an oil lamp, the remains of some fruit and the rolled-up lead sheet. It was placed underneath a clay pot and the upper body was facing down.Footnote 61 From archaeological observation it seems that the figurine had been thrown into the ditch and that the damage and position were not intentional. However the missing arms could suggest a ritual mutilation.

(b) Raetia (F4–F7)

Eining (Bavaria)/Abusina/unknown context

Two figurines have been published by Spindler.Footnote 62 Unfortunately their archaeological context is unknownFootnote 63 and their dating is indeterminate.

F4 (fig. 1c). The first figurine (12.1 cm tall) is made of clay. The body is stocky with a fairly rectangular head. The neck is thick and the figure has no shoulders. The nose and eye cavities seem to have been roughly shaped together. The eyes and mouth are depicted by holes. An additional clay strip stuck over the forehead indicates hair. The figurine presents exaggerated masculine genitals. The arms and legs are partially missing. It has been pierced in the chest, belly, forearms and legs.

F5 (fig. 1d). The second clay figurine represents only the upper part of a body. The dimensions are not given. Holes probably made with a fine tip depict the eyes and mouth. The nose has been made using simple finger pressure on the clay. The head is circular but irregular. The forearms are missing. The figurine has been pierced in the chest. It presents damage perhaps due to both natural drying of the clay and ritual piercing but the manufacture is similar to figurine F4.

Straubing (Bavaria)/Sorviodurum/unknown context

Two hollow figurines containing seeds were recovered from the site.Footnote 64 They were studied by Spindler who suggested that they were linked to a mystery cult and may have been used as rattles;Footnote 65 their appearance and manufacture, however, resemble magic dolls. Unfortunately no details about the species of the seeds have been provided. Nevertheless comparison can be made with a magic doll which contained a spell written on a papyrusFootnote 66 and with seven wax figurines from Rome shaped around a bone.Footnote 67 There might also be a connection with magic spells giving instructions (magical papyri).Footnote 68 Unfortunately the archaeological context of these two dolls is unknown, so that no further information can be provided about deposition and dating.

F6 (fig. 1e). The first figurine (10.5 cm tall) has been roughly shaped, with a hollow body containing the seeds. Finger-prints can still be seen. The head is irregular but its awkward form may include ears. The nose and forehead are prominent and thus seem to have been shaped together using finger pressure. The eyes are depicted merely by two small holes made in the lower part of the curved forehead. The neck has been coarsely shaped. The arms are short and their shaping has formed a hollow chest. The right leg is missing and the left is only partially preserved. The figurine also has prominent male genitals. Hollow lines at the hips suggest that the doll has been constricted by some kind of twine, while it has been clearly pierced under the chest.

F7 (fig. 1f). The second figurine (13.9 cm tall) has also been roughly shaped. The head is circular and the eyes are depicted by just two small holes made on the lower part of the curved forehead. A curved notch indicates the mouth. The neck is thick but the shoulders have been shaped. The nose and eye cavities are visible and are the result of quick finger modelling of the clay. The legs are short and the arms are partially missing. The left hip is strongly marked. The figurine presents prominent genitals and its masculine character can be recognised. A notch crosses the chest as if the ritual piercing has missed but another hole can be seen on its left side.

(2) AUSTRIA: UPPER PANNONIA (F8)

Petronell/Bad Deutsch-Altenburg (Lower Austria)/Carnuntum/sanctuary

The upper part of a pierced clay figurine was found during the excavation of the ancient sanctuary of Jupiter Heliopolitanus in the eastern part of the canabae legionis. It was published by Gassner.Footnote 69 The archaeological context dates the piece to the last quarter of the second century.

F8 (fig. 1g). This figurine (5.8 cm tall) has an awkward-shaped head with a thick neck. The transition between the head and the shoulders is not very well shaped. The skull is elongated and flattened on both sides. The upper part of the skull presents stripes with wavy edges which may indicate long curly hair. The face is triangular. The nose and eyes are roughly indicated and have probably been shaped using finger pressure on the clay. A small irregular ball of clay seems to be present inside the circular hole which represents the mouth. In this case a comparison can be made with defixiones carried out to ensure a victory in a legal dispute, many of which sought to incapacitate adversaries (witnesses and lawyers) by altering their oratorical skills or their memory.Footnote 70 The doll has been pierced below the neck. The figurine comes from a late second-century well filled with ritual remains. This large pit contained charcoal, crushed stones, architectural and sculpted fragments, and wall-painting fragments, so it is not unlikely that its filling marks the destruction of a building. At the same time the possibility that this well had some ritual purpose cannot be ruled out. It also contained two major groups of finds: an homogeneous assemblage of terra sigillata (with a high proportion of drinking vessels, notably bowls of type Drag. 37, Cup Ludowici VM and Drag. 33, and a few larger bowls and incense burners)Footnote 71 as well as cups and numerous animal bones (selected parts of cattle and a high percentage of bird bones).Footnote 72 The exact location of the figurine and the type of disposal were not documented.Footnote 73 The discovery of a single head within the rubble may recall the decapitation of some figurines such as the example from Attica which was deliberately decapitated, folded and pierced by two bronze nails.Footnote 74

(3) BRITAIN: BRITANNIA (F9)

Fishbourne (Sussex)/villa

A lead figurine was found by Kenny during excavations conducted in the neighbourhood of the Roman Palace.Footnote 75 It has been interpreted as a magic doll.Footnote 76

F9 (fig. 1h). This figurine (4.3 cm tall) has been roughly made, but a very thin tip was used to depict the hair and eyes and a broader tool to define the buttocks. Its forehead is prominent and it has no mouth. The end of its right leg and its forearms and hands are missing. The damage to the upper limbs may be due to twisting. The left leg is longer than the right and bent under the body at the knee. There are no indications concerning its gender but notches on the face could represent a beard.Footnote 77

The archaeological context of the figurine — an excavation spoil heap — provides no further information about the depositFootnote 78 but evidence collected from excavations elsewhere at Fishbourne (several votive finds, amulets and remains linked to a cult devoted to Cybele and Attis)Footnote 79 offers many clues to ritual activities. Moreover discussion of Building 3 (located about 100 m west of the area excavated in 1992) points to a religious/ritual explanation.Footnote 80

(4) FRANCE

(a) Gallia Belgica (F10–F11)

Reims/Durocortorum (Champagne-Ardennes)/cellar of large building

Two third- to fourth-century figurines were found in the cellar of a late Roman buildingFootnote 81 during excavations by INRAP on Rue Saint-Symphorien. These revealed the presence of three magic deposits in the cellar of a large building (a domus or a schola) constructed in the late second or early third century and destroyed at the beginning of the fourth century.Footnote 82 These deposits were placed along the north wall of the cellar. One included the figurines.Footnote 83

F10 (fig. 1i). The first figurine has been shaped from clay (8.4 cm tall). It seems that the lower part of its stocky body has been cut. The head is fairly rectangular. The neck is thick but the shoulders have been shaped. The arms, chest, belly and details of the face have been severely damaged. Needles or nails have deeply pierced the figurine all over the body (neck, under the arms, belly and pelvis). It was placed in a box made from two terra sigillata bowls (fig. 1j).Footnote 84 The upper bowl was covered by seven small balls of a dark paste (or wax), which almost certainly had a ritual function. An iron key was placed beside the upper bowl and it is presumably linked to the symbolic ‘sealing’ of the curse ritual.

F11. A small pile of clay seems to represent the remains of another doll which has presumably disintegrated. The discovery can be compared with the remains of one of the three figurines at Mainz. The location of the magic deposits — a cellar — cannot simply be interpreted as a domestic context. In Italy and Gaul, some cellars were also places for ritual activities as is demonstrated by certain private oratories located in building basements.Footnote 85

(b) Gallia Lugdunensis (F12)

Piriac-sur-Mer (Pays de la Loire)/villa

A lead figurine was found in 2004 while INRAP was carrying out an archaeological evaluation prior to the construction of a Business Park.Footnote 86

F12 (fig. 1k). The lead figurine (about 13 cm tall, 5 cm wide and 450 g in weight) is crude. Its regular thickness and crude shape suggest that it has been roughly cut from a flat piece of lead. Both hands are twisted. The left arm is bent above the chest. The right arm is twisted behind the top of its right shoulder as if bound. It is brokenFootnote 87 and the damage is presumably due to deliberate twisting. The legs are bent. The left leg is shorter and incomplete. A notch can be seen in the middle. The right leg is bent at the knee and bound. Details of the eyes, mouth, chest and hair seem to have been engraved using a tool with a thin tip and perhaps stamped with a circular object. Its exaggerated genitals are flattened above the abdomen.

The archaeological context is well documented. It was found during an evaluation trench dug in the west wing of the pars rustica of the villa which was occupied by a wine-press, within late archaeological levels which also produced finds such as iron nails, a fourth-century terra sigillata sherd and nine bone fragments, one of which was burnt. Even though the context provided no other evidence of ritual activities, the appearance of the figurine suggests it was magic.Footnote 88 The presence of the figurine next to a wine-press raises the question of a potential link between the device and the curse ritual. Was it supposed to be ground up with other magic material and other dolls? Was the rotation of the press symbolically used in the ritual? Magic figurines completely twisted around, with head and feet or head twisted to the side, are not uncommon.Footnote 89 Was wine required in the ritual? Spells indicating the use of wineFootnote 90 and maybe grinding are frequent in the Greek Magical Papyri.Footnote 91 Nevertheless, it is difficult to prove the connection, as the figurine comes from the destruction levels, hence its deposition need not necessarily be linked with the use of the press and the wine-making process. Moreover, its association with bones, nails and terra sigillata could simply reflect the abandonment of the item and not necessarily any ritual link.

CONCLUSION

This survey has revealed that magic figurines are often found in important places such as exceptional villas or provincial capitals. The presence of magical objects in the sophisticated palace of Fishbourne or in a major building (Reims or Piriac) might indicate that magic was performed at all levels of society and not only in the most marginal groups. This is what Tacitus seems to indicate in the case of Germanicus, general and member of the Julio-Claudian dynasty. In a.d. 19, near Antiochia, Germanicus was informed that he had been cursed. The information was based on lead curse-tablets with his name on them, human remains and partly burned bones recovered in the walls and the floor of the house in which he was lodged.Footnote 92 It is also what seems to be confirmed by the curse tablets from Britannia which are motivated by the theft of valuable personal properties.Footnote 93 Moreover a defixio written on three different sheets found at Ampurias (Spain) seems to complete the testimony of Tacitus.Footnote 94 The issue is a border dispute between the Olossitani and the Indicetani. The curse targets the legatus Augusti, his assistant and the lawyer defending the Indicetani.Footnote 95

The survey also demonstrates that the magic practised in the West has clear parallels with the ritual-magical procedures known from the Greek Magical Papyri as performed in the Mediterranean basin. The places where figurines come from were in the main propitious to trade and exchange and exposed to outside influences. Through mapping, the survey indicates that some of these places provide evidence of oriental cults and the presence of the military, as at Carnuntum and Mogontiacum. It also underlines the role of private locations in magic, such as the cellar in Reims, in addition to the usual places for ritual (cemeteries and temples). Nevertheless, they still show an indisputable connection with chthonic deities. The dating of the Western figurines is centred on the second century, though the most recent figurine possibly dates from the fourth century. It is also interesting to note that the dating matches a period of progressive change, notably in the religious domain.Footnote 96

Footnotes

1 e.g. Tupet Reference Tupet1976.

2 Audollent Reference Audollent1904; Preisendanz Reference Preisendanz1973–1974; Betz Reference Betz1992; Bonner Reference Bonner1950; Kotansky Reference Kotansky1994; Jordan Reference Jordan1985; Reference Jordan2000; Faraone Reference Faraone1991a. Other apotropaic symbols can be mentioned including the medusa, owl and fascinum (Johns Reference Johns1982; Dunbabin and Dickie Reference Dunbabin and Dickie1983; Trombetta Reference Trombetta1999; Bailliot Reference Bailliot2010, 28–69).

3 Many scholars suggest that lead was used for curses because of its properties: darkness and coldness. Some tablets support this hypothesis such as a fourth-century b.c. defixio from Attica: ‘Let Pherenikos be bound before Hermes Chthonios and Hekate Chthonia … just as this lead is worthless and cold’ (Audollent Reference Audollent1904, n. 107, translated by Faraone Reference Faraone1999, 112). It has to be emphasised that other kinds of writings were engraved on lead (e.g. lead labels with written texts and commercial letters on lead-plaques) and that wax for example — which is not described as a specific magical material — was used for defixiones as suggested by Ovid (Amores 3.7.28–30): ‘Do charms and herbs hurt my poor self now, some witch transfixes my name in scarlet wax (defixit nomina cera) and sticks fine needles right into my liver?’ (translation: Kline Reference Kline2001). See also the verb figo in Ernout and Meillet Reference Ernout and Meillet1939, 234. Lead was also used to make cult statuettes (e.g. a small lead plaque depicting a goddess from Fishbourne Roman Palace: Bailliot and Symmons Reference Bailliot and Symmons2012, 252, fig. 6a. For other examples see Green Reference Green1978). Magicians used both materials (lead and wax) certainly because they were easy to engrave. The magical properties of lead — darkness and coldness — which cannot be totally rejected are only known through late exegesis (Charvet and Ozanam Reference Charvet and Ozanam1994, 18).

4 Luck Reference Luck1985; Graf Reference Graf1994; Ogden Reference Ogden2002.

5 Audollent Reference Audollent1904.

6 They can also be found in domestic contexts as at Kempten in south-west Bavaria (Solin Reference Solin1968, n. 10), Leicester (Tomlin Reference Tomlin2008) and Chartres (Viret et al. Reference Viret, Lambert, Stüber, Stifter and Repanšek2013).

7 Jordan Reference Jordan1985, n. 1; Del Amor Lopez-Jimeno Reference Del Amor López Jimeno1991.

8 e.g. defixiones from Apamea: SEG 34, 1437–8; defixio from Antiochia: Jordan Reference Jordan2000, n. 102; defixio from Amathus: Aupert Reference Aupert2008, 370–87. There are fewer such late agonistic inscriptions and all are related to the practice of chariot racing which continued at least until the seventh century (Liebeschuetz Reference Liebeschuetz2003, 219).

9 Johnston Reference Johnston2003.

10 e.g. Schilling Reference Schilling1954.

11 Versnel Reference Versnel1976, 370.

12 Versnel Reference Versnel1991a. Therein Versnel emphasises that ‘magic does not exist, nor does religion; what do exist are our definition of these concepts’ (ibid., 177). For a review of the strengths and weaknesses of the definitions given to magic see Meggitt Reference Meggitt2013.

13 e.g. Apuleius who was accused of using magic to gain the attentions of a rich widow (Graf Reference Graf1994, 79ff.; ibid., 50–1 for another example). Examples of papyri burning can also bear witness to the illegality of magic, as in the case of the burning of magic books in a.d. 13 (Suetonius, Augustus 31.1): ‘After he finally had assumed the office of pontifex maximus on the death of Lepidus (for he could not make up his mind to deprive him of the honour while he lived), he collected whatever prophetic writings of Greek or Latin origin were in circulation anonymously or under the names of authors of little repute, and burned more than two thousand of them, …’ (translation: Rolfe Reference Rolfe1913). The death of Germanicus was attributed to a magic curse (see Tacitus, Annals 2.69; translation: Jackson and Moore Reference Jackson and Moore1931).

14 Scheid Reference Scheid2005, 127.

15 e.g. of constraint in magical texts (PGM iv, 2089–93). Blackmail was also used by magicians (Graf Reference Graf1994, 249–52).

16 Bailliot Reference Bailliot2010, 150–7. The etymology of the word sacrilegium (from sacer: ‘sacred, holy’ and legō: ‘gather; take, steal’) implies that sacrilege means ‘to steal the possessions of the gods from the temples or dwelling places’. Cf. ‘To make my meaning clearer, I will follow my usual practice and quote a familiar example. A man who has stolen private money from a temple is accused of sacrilege’. Also: ‘It is sacrilege to steal something sacred’ (Quintilian, Institutes 7.3; translation: Butler Reference Butler1920). Sacrilegium is illustrated by a few examples discussed by Scheid (Reference Scheid1981, 137–42), such as magic formulae which accuse the victims of having committed an impiety (PGM iv, 2441–2621; iv, 2622–2700; vii, 593–619).

17 Versnel Reference Versnel1987, 9.

18 e.g. Witteyer Reference Witteyer2005; Piranomonte Reference Piranomonte2005; Reference Piranomonte2012; Bailliot Reference Bailliot2010; Bevilacqua et al. Reference Bevilacqua, Colacicchi and Giuliani2012; Wilburn Reference Wilburn2013.

19 Papyri Graecae Magicae (abbreviated PGM) contain spells, formulae, hymns, rituals and notably various ways to curse enemies. Greek and Demotic papyri recovered in Egypt are dated between the second century b.c. and the fifth century a.d. They were also used in other provinces of the Roman Empire (e.g. Betz Reference Betz1992, xli). The body of these documents was first translated by Karl Preisendanz and published from 1928 (Preisendanz Reference Preisendanz1973–1974).

20 Research which notably shows local variations in the ritual (Mees Reference Mees2009; Gordon and Marco Simón Reference Gordon and Marco Simón2010; Bailliot Reference Bailliot2010; Wilburn Reference Wilburn2013) emphasises the discrepancy between the reality and the notions of ‘subversive magic’ inherited from the nineteenth century.

21 Graf Reference Graf1994. This wrong assumption is based on the idea of sympathetic magic inherited from James George Frazer's work. See also Versnel Reference Versnel and Graf1998.

22 Faraone Reference Faraone1991a; Versnel Reference Versnel1988, 289–90; Ogden Reference Ogden2002, 245–6; Del Amor López Jimeno Reference Del Amor López Jimeno2010. Some of those figurines present prominent genitals or show quite a high level of realism (e.g. Du Bourguet Reference Du Bourguet1975; Reference Du Bourguet1980; Kambistis Reference Kambistis1976), while the majority are roughly shaped. Most were used for erotic purposes but some for other ends such as legal disputes, as the defixio found with figurine F3 at Mainz suggests. This curse, only indicating the personal name of the victim (‘Trutmo Florus, son of Clitmo’), could be a judicial prayer (curse against thieves) (Versnel Reference Versnel1991b; Reference Versnel2010). Nine small lead figurines depicting race-horses recovered at Antiochia (Seyrig Reference Seyrig1935, 48–50) demonstrate that magic also targeted animals in order to overwhelm their owners.

23 Greek mainland and the islands, Sicily, Italy, Egypt, North Africa, Palestine, Asia Minor and Black Sea areas (Faraone Reference Faraone1991a, 200–5).

24 Faraone Reference Faraone1991a.

25 e.g. Del Amor López Jimeno Reference Del Amor López Jimeno2010. Figurines are not always found together with lead sheets but we may presume that spells may also have been written on biodegradable materials which have not survived. I would like to thank Stéphane Sindonino and Michaël Brunet for allowing me to study the archaeological evidence from Reims, and the librarians who gave me access to unpublished sources.

26 e.g. the description by Horace of a magic rite carried out within the cemetery of the Esquiline at Rome (Satires 1.8.23–50): ‘I've seen Canidia myself, wandering barefoot /With her black robe tucked up, and disheveled hair, /Howling with the elder Sagana: pallor making them /Hideous to view. They scraped at the soil with their nails, /Then set to tearing a black lamb to bits with their teeth: /The blood ran into the trench, so they might summon /The souls of the dead, spirits to give them answers. /There was a woollen doll there, and another of wax: /The wool one was larger to torment and crush the other. /The wax one stood like a suppliant, waiting slave-like /For death’ (translation: Kline Reference Kline2005).

27 Graf Reference Graf1994, 142–51 and 232–3; Scheid Reference Scheid2005; Bailliot Reference Bailliot2010, 19 and 72–3. Anthropologists describe performative rituals as formulae which strengthen, modify or create a situation by virtue of simply being uttered (e.g. the wedding formula ‘I now pronounce you husband and wife’ which creates the legal union).

28 Faraone Reference Faraone1991b, 8.

29 Wünsch Reference Wünsch1902.

30 Faraone Reference Faraone1999, 58.

31 / ardeat Vettia quam peper[it Optata ---] / [a]moris et desideri‹i›(!) m[ei causa ---] / [------] (Audollent Reference Audollent1904, n. 266). It should be noted that Versnel (Reference Versnel and Graf1998, 146–64) distinguishes binding spells containing selections of body parts which indicate that the victim should be restrained in his actions (e.g. transfixing a tongue in the context of a legal dispute) from defixiones with descriptions of tortures which are not directly relevant to the social context of the curse. In the latter case, he argues that the abundant descriptions of tortures were made to exert an irresistible rhetorical pressure upon the addressee. The same assumption can be proposed for the figurines. The action of burning them (which can correspond to fire and to fever) does not necessarily indicate an erotic curse.

32 Faraone Reference Faraone1999, 66.

33 Audollent Reference Audollent1904, nn. 111–12.

34 … quo/modi(!!) hic catellus aversus / est nec surgere potest{i} / sic nec illi sic tra‹n›sp‹i = E›cti(!!)(?) sin[t] / quomodi(!!) ille / quomodi(!) in hoc m[o]n‹u = I›m‹e = O›nt‹o› ani/malia o‹b = M›mutuerun‹t› nec surge/re possun‹t› nec illi mut[i](?) / (translated by Faraone Reference Faraone1999, 66–7).

35 [in]imicos(!) /Atracatetracati gal/lara precata egdarata / hehes celata mentis ablata / {[et] a‹d = T› Proserpinam hinc a[beant]} // (Audollent Reference Audollent1904, n. 241; translated by Faraone Reference Faraone1999, 101).

36 For various examples see Faraone Reference Faraone1991a.

37 Jordan Reference Jordan and Humphrey1988, n. 3. The fetters were also replaced by snakes on figurative defixiones at Rome (Wünsch Reference Wünsch1898). For other examples and other potential comparisons see Bailliot Reference Bailliot2010, 112–27.

38 Bailliot Reference Bailliot2010, 113; Wilburn Reference Wilburn2013, 74–5.

39 Two fifth-century wax figurines from Egypt — recovered with a love spell — depicted in an embrace are of note (Gager Reference Gager1992, 15, 101–6; Wilburn Reference Wilburn2013, 135–6). This discovery is the only example showing figurines in such a position. It enhances the formulae of the associated defixio.

40 Cumont Reference Cumont1913. The left leg of the figurine is folded. The hands are tied behind the back. The head is bowed and the hair is wrapped. The box is 11.5 by 0.55 cm (3.5 cm high). The doll is 11 cm tall.

41 Piranomonte Reference Piranomonte2005; Reference Piranomonte2010; Polakova and Rapinesi Reference Polakova, Rapinesi and Piranomonte2002.

42 Piranomonte Reference Piranomonte2010, 206.

43 For funerary rituals see Scheid Reference Scheid2005.

44 Bailliot Reference Bailliot2010, 72–6; Wilburn Reference Wilburn2013, 91–2.

45 Audollent Reference Audollent1904, n. 139. ‘Just as this dead person who is buried here can neither speak nor talk, so may Rhodine die as far as Marcus Licinius Faustus is concerned and not be able to speak nor talk. As this dead person is received neither by gods nor humans, so may Rhodine be received by Marcus Licinius and have as much strength as the dead man who is buried here. Dis Pater, I entrust Rhodine to you, that she be always hateful to Marcus Licinius Faustus. Also Marcus Hedius Amphio. Also Gaius Popillius Apollonius. Also Vennonia Hermiona. Also Sergia Glycinna’ (translation: Beard et al. Reference Beard, North and Price1998, 268).

46 Audollent Reference Audollent1904, n. 98.

47 Ernoult and Meillet Reference Ernout and Meillet1939, 194–5.

48 Bailliot Reference Bailliot2010, 105.

49 Bailliot Reference Bailliot2010, 104–12; Wilburn Reference Wilburn2013, 91–2.

50 Witteyer Reference Witteyer2005. The finds are kept in the Taberna archaeologica at Mainz (General Direktion Ländes Archäologie).

51 Witteyer Reference Witteyer2005, 111.

52 Witteyer Reference Witteyer2005, 109.

53 Wilburn Reference Wilburn2013, 132–6.

54 That suggests it could have been partially baked as was figurine F1.

55 Witteyer Reference Witteyer2005, 111.

56 Blänsdorf Reference Blänsdorf2005, 13 n. 1; Reference Blänsdorf2010, 156 n. 10. The text is: ‘Trutmo Florus, son of Clitmo’.

57 Witteyer Reference Witteyer2005, 114.

58 Prodigium was a sign about the future that came from the gods in the form of an incident or an abnormal phenomenon announcing an impending calamity over a whole community. To avert disaster, sacrifices or propitiating actions were performed to appease the concerned god(s).

59 Delcourt Reference Delcourt1986. Cicero (On the Laws 3.8.19) mentions that the Law of the Twelve Tables notes that a deformed child must be killed swiftly: ‘Cito necatus tamquam ex XII tabulis insignis ad deformitatem puer’ (translation: Nisard Reference Nisard1864).

60 A case occurred in Latium in 210 b.c. and is related by Livy (The History of Rome 27.37.5): ‘After the people's minds had been freed from superstitious fears, they were again disturbed by intelligence that an infant had been born at Frusino as large as a child of four years old, and not so much an object of wonder from its size, as that it was born without any certain mark of distinction whether it was male or female, which was the case two years before at Sinuessa. Aruspices, called in from Etruria, declared this to be indeed a foul and ill-omened prodigy, which ought to be removed out of the Roman territory, and, being kept far from coming in contact with the earth, to be plunged into the deep. They shut it up alive in a chest, and carrying it away, threw it into the sea’ (translation: Edmonds Reference Edmonds1850).

61 Witteyer Reference Witteyer2005, 111.

62 Spindler Reference Spindler1984, 113, fig. 96. Unknown location.

63 The site located near the Danube, 30 km from Regensburg (Castra Regina), included a fort, a villa and two sanctuaries (Schmid Reference Schmid1910; Kellner Reference Kellner1971).

64 This site, located in Lower Bavaria on the bank of the Danube, included a fort and a villa (Walke Reference Walke1965).

65 Spindler Reference Spindler1984, 114, figs 97, 98. Unknown location.

66 Ogden Reference Ogden, Ankarloo and Clark1999, 14, fig. 5.

67 Piranomonte Reference Piranomonte2005, 101; Reference Piranomonte2010, 206–7.

68 Some Greek Magical Papyri provide instruction on how to fashion a figurine, as for example a Hermes with a hollow bottom in PGM iv, 2359–72: (Business spell) ‘Take orange beeswax and I the juice of the aeria plant and of ground ivy and mix them and fashion a figure of Hermes having a hollow bottom, grasping in his left hand a herald's wand and in his right a small bag. Write on hieratic papyrus these names, and you will see continuous business: I “CHAIŌCHEN OUTIBILMEMNOUŌTH ATRAUICH. Give income and business to this place, because Psentebeth lives here”. Put the papyrus inside the figure and fill in the hole with the same beeswax. Then deposit it in a wall, at an inconspicuous place, I and crown him on the outside, and sacrifice to him a cock, and make a drink offering of Egyptian wine, and light for him a lamp that is not colored red’ (translation: Betz Reference Betz1992, 81). For other instructions see: PGM cxxii, 1–43 (Charm to inflict illness) and iv, 296–466 (Wondrous spell for binding a lover).

69 Gassner Reference Gassner2010. Precise location unknown. The site was a Roman army camp settled on the Danube in Noricum. After the first century it became the capital of Upper Pannonia. The site was dug from 1978 to 1991 prior to a housing development. The pit from where the figurine comes was dug in 1979 (Gassner et al. Reference Gassner, Steigberger and Tober2011, 129).

70 For various examples see Audollent Reference Audollent1904.

71 Gassner et al. Reference Gassner, Steigberger and Tober2011, 147. The assemblage (Drag. 37, Lud. VM, Drag. 33, Drag. 31, Drag. 32, Drag. 43) dates the filling of the pit between a.d. 170/80 and 210/20. two coins have also been collected: a sestertius (a.d. 163.) and an as (a.d. 175/6).

72 Gassner et al. Reference Gassner, Steigberger and Tober2011, 146.

73 Gassner Reference Gassner2010, 224, fig. 3.

74 Wünsch Reference Wünsch1902, 26–31.

75 Kenny Reference Kenny and Woodward1992a. The archaeological remains recovered in 1961 are very well known and were excavated by Cunliffe. The large villa was built in the first century and burnt down around a.d. 270 (Cunliffe Reference Cunliffe1971). It is the largest villa in Britannia and Northern Europe.

76 Bailliot and Symmons Reference Bailliot and Symmons2012. The object is held in the care of Sussex Archaeological Society at Fishbourne Roman Palace.

77 Bailliot and Symmons Reference Bailliot and Symmons2012, 249.

78 Kenny Reference Kenny and Woodward1992a. Its context indicates a late second-century date and metalworking activities (Kenny Reference Kenny and Woodward1992b, 34). These would have supplied the nearby palace (decorative and functional items) during its renovation. The industrial activity was probably linked to this work.

79 Cunliffe Reference Cunliffe1971; Manley and Rudkin Reference Manley and Rudkin2003b, 188; Rudling Reference Rudling and Rudling2008; Bailliot and Symmons Reference Bailliot and Symmons2012, 252.

80 Manley and Rudkin Reference Manley and Rudkin2003a, 4, 20–9; Manley Reference Manley2000, 111.

81 Reims/Durocortorum became the capital of Gallia Belgica during the reign of Augustus.

82 Sindonino Reference Sindonino2011; forthcoming.

83 Bailliot Reference Bailliot2010, 96–7. The finds are currently held in the care of INRAP Champagne-Ardennes. A complete publication of the magic deposits is forthcoming (Sindonino et al. forthcoming).

84 Analyses of the finds carried out by INRAP are underway.

85 Santrot Reference Santrot, Baratte, Joly and Béal2007, 82; Joly Reference Joly2010; Reference Joly2012; Mauduit and Lambert Reference Mauduit, Lambert, Rivet and Saulnier2008.

86 Letho-Duclos Reference Letho-Duclos2004. The doll is held in the care of INRAP Loire-Atlantique. Full excavation in 2005, following archaeological evaluation, has revealed the remains of a Gaulish settlement followed by a Roman villa extending over 3.5 ha. The settlement, established during the first century, developed significantly during the Roman period (Hervé-Monteil Reference Hervé-Monteil2010). Other remains indicate that the site was still active during the seventh century. During the first century, the site produced purple from nucella lapillus used for dyeing textiles. This production probably finished when the villa and its annexes, which were devoted to stockbreeding activities, were constructed. The settlement changed during the second century. New buildings were erected, including a warehouse (for agricultural goods), a house with a forecourt and private baths. Between the second and fourth century, the warehouse became a winery and a wine-press was added.

87 Letho-Duclos Reference Letho-Duclos2004, fig. 37. The figurine has been restored to its initial state by the Arc'Antique laboratory in Nantes.

88 It recalls magic dolls studied by Faraone. It is flat, similar to a fourth-century doll (9 cm tall and 5 cm wide) found at Carystus on the island of Euboea in Greece (Faraone Reference Faraone1991a, 202 n. 15, pl. 11) and to 16 bound figurines from Tell Sandahannah (Palestine) cut in lead plaques (height around 5.08 and 7.62 cm) (Clermont-Ganneau Reference Clermont-Ganneau1901, 156). It is also similar to a couple of figurines recovered at Volubilis (Souville Reference Souville1957). The Piriac figure is neither an amulet used to moderate sexual desire nor the representation of a local fertility god as first suggested (Hervé-Monteil Reference Hervé-Monteil2010, 52; Bertrand and Cartron Reference Bertrand and Cartron2010, 30).

89 Versnel Reference Versnel and Graf1998, 289–90.

90 e.g. PGM i, 2359–72; iv, 1716–1870; xi, 636–69; xiv, 366–5.

91 Versnel Reference Versnel1988.

92 ‘After this, he left for Seleucia, awaiting the outcome of the malady which had again attacked Germanicus. The cruel virulence of the disease was intensified by the patient's belief that Piso had given him poison; and it is a fact that explorations in the floor and walls brought to light the remains of human bodies, spells, curses, leaden tablets engraved with the name Germanicus, charred and blood-smeared ashes, and others of the implements of witchcraft by which it is believed the living soul can be devoted to the powers of the grave. At the same time, emissaries from Piso were accused of keeping a too inquisitive watch upon the ravages of the disease’ (Tacitus, Annals 2.69; translation: Jackson and Moore Reference Jackson and Moore1931).

93 Tomlin Reference Tomlin1988; Reference Tomlin, Woodward and Leach1993; Bailliot Reference Bailliot2010, 143–7.

94 Almagro Reference Almagro1952, 114–16; Solin Reference Solin1968, nn. 26, 27, 28; Gager Reference Gager1992, n. 52.

Maturus proqura/tor Augusti consi/lium legati/ legati Indiceta/norum [consilium] Indiceta/ (norum)

Olossitan[i] / Titus Aurelius Fulvus lega/tus Augusti /Rufus legatus Au/gus[ti].

Consilium M. Fulvi / legati Olossi/tani Campanus/ [Fid]enti[nus at…] -

Fulvus legatus Au/gusti Rufus legatus/ Augusti Maturus/ proqurator Augusti/ legati atvocati [Indi]cetanoru[m].

[Oloss]itani / Sempronius / Campanus Fi/dentinus at/versari / M. . . .inique/ Nem. . . nt… -

[Ful]vus legatus/ [Aug]usti Rufus lega/tus Augusti Matu/[ru]s pro[cu]r[at]or/ [Au]gusti [c]ons[i]lium / [legati] atvocati [Indice]/t[ano]ru[m]

1. (side A) Maturus, Augustan procurator; councilor of the

legate, (that is) the councilor of the legate for the Indicetani;

(the advocate ?) for the Indicetani

(side B) The Olossitani, Titus Aurelius Fulvus, Augustan

legate; Rufus, Augustan legate

2 (side A) Councilor of M. Fulvus, the Olossitani, Campanus

Fidentius …

(side B) Fulvus, Augustan legate; Rufus, Augustan legate

Maturus, Augustan procurator, councilor of the legate;

advocates of the Indicetani

3 (side A) Sempronius Campanus Fidentinus of the Olossitani.

.. oppose me unfairly …

(side B) Fulvus, Augustan legate; Rufus, Augustan legate;

Maturus, Augustan procurator; councilor of the legate;

advocates of the Indicetani

(Translated by Gager Reference Gager1992, n. 52.)

95 Maturus, the procurator, seems to be Marius Maturus who was procurator Alpium marilimarum in a.d. 69 (ibid.): ‘The Governor of the Maritime Alps at that time was Marius Maturus. He summoned the inhabitants, whose fighting strength was ample, and proposed to resist at the frontier the Othonians’ invasion of the province. But at the first engagement the mountaineers were cut down and dispersed. They had assembled in random haste; they knew nothing of military service or discipline, nothing of the glory of victory or the disgrace of flight’ (Tacitus, Histories 2.12; translation: Hamilton Fyfe Reference Hamilton Fyfe1912).

96 e.g. Turcan Reference Turcan1992; Frankfurter Reference Frankfurter2000.

References

BIBLIOGRAPHY

Almagro, M. 1952: Monografías ampuritanas griegas, ibéricas y latinas, BarcelonaGoogle Scholar
Audollent, A. 1904: Defixionum tabellae quotquot innotuerunt tam in Graecis Orientis quam in totius Occidentis partibus praeter Atticas in corpore inscriptionum Atticarum editas, Thèse de Doctorat d'État, ParisGoogle Scholar
Aupert, P. 2008: ‘Hélios, Adonis et magie: les mystères d'une citerne d'Amathonte’, Bulletin de Correspondance Hellénique 132.1, 347–87CrossRefGoogle Scholar
Bailliot, M. 2010: Magie et sortilèges dans l'Antiquité romaine. Archéologie des rituels et des images, ParisGoogle Scholar
Bailliot, M., and Symmons, R. 2012: ‘Note from the Roman Palace at Fishbourne (Sussex): a Roman magic lead figurine?’, Britannia 43, 249–60CrossRefGoogle Scholar
Beard, M., North, J., and Price, S. 1998: Religion of Rome. A Sourcebook, vol. 2 (5th edn), CambridgeGoogle Scholar
Bertrand, I., and Cartron, I. 2010: ‘Figurines en plomb du sanctuaire de “La Chapelle” à Jau-Dignac-et-Loirac (Gironde, F)’, Instrumentum 32, 29–31Google Scholar
Betz, H.D. 1992: The Greek Magical Papyri in Translation: Including the Demotic Spells, Chicago/London (original edition, 1986, Chicago/London)Google Scholar
Bevilacqua, G., Colacicchi, O., and Giuliani, M.-R. 2012: ‘Tracce di ousia in una defixio dalla via Ostiense: un lavoro multidisciplinare’, in Piranomonte and Marco Simón 2012, 229–36Google Scholar
Blänsdorf, J. 2005: ‘The curse tablets from the Sanctuary of Isis and Mater Magna in Mainz’, Revista Internacional de Investigación sobre Magia y Astrología Antiguas 5, 11–26Google Scholar
Blänsdorf, J. 2010: ‘The defixiones from the Sanctuary of Isis and Mater Magna in Mainz’, in Gordon and Marco Simón 2010, 141–90CrossRefGoogle Scholar
Bonner, C. 1950: Studies in Magical Amulets. Chiefly Graeco-Egyptian, Ann ArborGoogle Scholar
Butler, H.E. 1920: Quintilian, Institutes, Loeb Classical Library, Cambridge, MAGoogle Scholar
Charvet, P., and Ozanam, A.-M. 1994: La Magie, textes traduits et présentés, ParisGoogle Scholar
Clermont-Ganneau, M. 1901: ‘L'envoutement dans l'Antiquité et les tablettes de Tell Sandahannah’, Revue d'Archéologie Orientale 4, 156–8Google Scholar
Cumont, F. 1913: ‘Figurines grecques d'envoûtement’, Comptes-rendus de l'Académie des Inscriptions et Belles-Lettres, 417–21CrossRefGoogle Scholar
Cunliffe, B. 1971: Fishbourne, a Roman Palace and its Garden, LondonGoogle Scholar
Del Amor López Jimeno, M. 1991: Las tabellae defixionis de la Sicilia griega, Classical and Byzantine Monographs 12, AmsterdamGoogle Scholar
Del Amor López Jimeno, A. 2010: ‘Una figurita de plomo hallada en Paros y otras figuritas de magia maléfica’, Revista Internacional de Investigación sobre Magia y Astrología Antiguas 10, 101–18Google Scholar
Delcourt, M. 1986: Stérilités mystérieuses et naissances maléfiques dans l'Antiquité classique, Bibliothèque de la Faculté de Philosophie et Lettres de l'Université de Liège 83 (original edition, 1938), ParisCrossRefGoogle Scholar
Du Bourguet, P. 1975: ‘Ensemble magique de la période romaine en Égypte’, La Revue du Louvre 25, 255–7Google Scholar
Du Bourguet, P. 1980: ‘Une ancêtre des figurines d'envoûtement percées d'aiguilles, avec ses compléments magiques, au Musée du Louvre’, Mémoires de l'Institut Français d'Archéologie Orientale du Caire 104, 225–38, figs 34–8Google Scholar
Dunbabin, K.M.D., and Dickie, M.W. 1983: ‘Invidia rupantur pectora, the iconography of phtonos/Invidia in Graeco-Roman art’, Jahrbuch für Antike und Christentum 26, 35–6Google Scholar
Edmonds, C. 1850: Titus-Livy, The History of Rome, LondonGoogle Scholar
Ernout, A., and Meillet, A. 1939: Dictionnaire étymologique de la langue latine, histoire des mots, ParisGoogle Scholar
Faraone, C.A. 1991a: ‘Binding and burying the forces of evil: the defensive use of “voodoo dolls” in ancient Greece’, Classical Antiquity 10/2, 165–205CrossRefGoogle Scholar
Faraone, C.A. 1991b: ‘The agonistic context of early Greek binding spells’, in Faraone and Obbink 1991, 3–32CrossRefGoogle Scholar
Faraone, C.A. 1999: ‘Curses and social control in the law courts of Classical Athens’, Dike: Revista di storia del diritto greco ed ellenistico 2, 99–121Google Scholar
Faraone, C.A., and Obbink, D. (eds) 1991: Magika Hiera. Ancient Greek Magic and Religion, New York/OxfordCrossRefGoogle Scholar
Frankfurter, D. 2000: ‘The consequences of Hellenism in late antique Egypt: religious worlds and actors’, Archiv für Religionsgeschichte 2, 162–94CrossRefGoogle Scholar
Gager, J.G. 1992: Curse Tablets and Binding Spells from the Ancient World, Oxford/New YorkCrossRefGoogle Scholar
Gassner, V. 2010: ‘Eine Zauberpuppe aus dem Heiligtum des Iuppiter Heliopolitanus in Carnuntum’, in ANODOS Studies of the Ancient World in Honour of Werner Jobst (8/2008), Trnava, 221–30Google Scholar
Gassner, V., Steigberger, E., and Tober, B. 2011: ‘Das Heiligtum des Iuppiter Heliopolitanus in Carnuntum. Überlegungen zu den älteren Kultbauten an der Ostseite, ihrer Ausstattung und den Mechanismen ihrer Aufgabe’, Carnuntum Jahrbuch, Zeitschrift für Archäologie und Kulturgeschichte des Donauraumes 2009–2011, 129–72Google Scholar
Gordon, R., and Marco Simón, F. (eds) 2010: Magical Practice in the Latin West. Papers from the International Conference held at the University of Zaragoza (30th September–1st October 2005), Leiden/BostonGoogle Scholar
Gordon, R., Joly, D., and Van Andringa, W. 2010: ‘A prayer for blessings on three ritual objects discovered at Chartres-Autricum (France, Eure-et-Loir)’, in Gordon and Marco Simón 2010, 487–518CrossRefGoogle Scholar
Graf, F. 1994: La Magie dans l'Antiquité gréco-romaine, idéologie et pratique, ParisGoogle Scholar
Green, M.J. 1978: A Corpus of Small Cult-Objects from the Military Areas of Roman Britain, BAR British Series 52, OxfordCrossRefGoogle Scholar
Hamilton Fyfe, W. 1912: Tacitus, Histories, OxfordGoogle Scholar
Hervé-Monteil, M.L. 2010: Piriac-sur-Mer (Loire-Atlantique), il y a 2000 ans: de la ferme gauloise à la villa gallo-romaine, Les Cahiers du Pays de Guérande 51 (numéro spécial)Google Scholar
Jackson, J., and Moore, C.H. 1931: Tacitus, Annals, Loeb Classical Library, Cambridge, MAGoogle Scholar
Johns, C. 1982: Sex or Symbols. Erotic Images from Greece and Rome, LondonGoogle Scholar
Johnston, S.I. 2003: ‘Review article: describing the undefinable: new books on magic and old problems of definition’, History of Religions 43.1, 50–4Google Scholar
Joly, D. (ed.) 2010: ‘L'attirail d'un magicien rangé dans une cave de Chartres/Autricum’, Gallia 67.2, 125–208Google Scholar
Joly, D. 2012: ‘La panoplie complète d'un magicien dans la cave d'une domus à Autricum (Chartres-France): C. Verius Sedatus, Carnute, gardien des divinités’, in Piranomonte and Marco Simón 2012, 211–23Google Scholar
Jordan, D.R. 1985: ‘A survey of Greek defixiones not included in the special corpora’, Greek Roman and Byzantine Studies 26, 151–97Google Scholar
Jordan, D.R. 1988: ‘New defixiones from Carthage’, in Humphrey, J.H. (ed.), The Circus and a Byzantine Cemetery at Carthage, vol. 1, Ann Arbor, 117–34Google Scholar
Jordan, D.R. 2000: ‘New Greek curse tablets’, Greek Roman and Byzantine Studies 41, 5–46Google Scholar
Kambistis, S. 1976: ‘Une nouvelle tablette magique du musée du Louvre’, Bulletin de l'Institut Français du Caire 76, 213–23Google Scholar
Kellner, H.J. 1971: Die Römer in Bayern, MunichGoogle Scholar
Kenny, J. 1992a: ‘A Roman lead figurine from the Fishbourne Westward House site’, in Woodward, A. (ed.), The Archaeology of Chichester and District 1992, Chichester, 72Google Scholar
Kenny, J. 1992b: ‘Fishbourne: Westward House’, in Woodward, A. (ed.), The Archaeology of Chichester and District 1992, Chichester, 32–7Google Scholar
Kline, A.S. 2001: Ovid, Amores, A.S. Kline's Poetry Archive: http://www.poetryintranslation.com/PITBR/Latin/AmoresBkIII.htmGoogle Scholar
Kline, A.S. 2005: Horace, Satires, A.S. Kline's Poetry Archive: http://www.poetryintranslation.com/PITBR/Latin/HoraceSatiresBkISatVIII.htmGoogle Scholar
Kotansky, R. 1994: Greek Magical Amulets: the Inscribed Gold, Silver, Copper and Bronze Lamellae, Part I, Published Texts of Known Provenance, Papyrologica Coloniensia 22.1, OpladenCrossRefGoogle Scholar
Letho-Duclos, Y. 2004: Diagnostic archéologique, Piriac-sur-Mer, ZA du Pladreau, la route de Guérande, unpub. Rapport de diagnostic (Opération n. 2004-041), Institut de recherches archéologiques préventives (INRAP)Google Scholar
Liebeschuetz, J. 2003: The Decline and Fall of the Roman City, OxfordGoogle Scholar
Luck, G. 1985: Arcana Mundi Magic and the Occult in the Greek and Roman Worlds, BaltimoreGoogle Scholar
Manley, J. 2000: ‘Measurement and metaphor: the design and meaning of building 3 at Fishbourne Roman Palace, Sussex’, Sussex Archaeological Collections 138, 103–13Google Scholar
Manley, J., and Rudkin, D.J. 2003a: Facing the Palace: Excavations in Front of the Roman Palace at Fishbourne (Sussex, UK), 1995–1999, Sussex Archaeological Collections 141Google Scholar
Manley, J., and Rudkin, D.J. 2003b: Facing the Palace: Excavations in Front of the Roman Palace at Fishbourne (Sussex, UK), 1995–1999, ADS Supplement. Sussex Archaeological Collections, 141 http://archaeologydataservice.ac.uk/archives/view/sac/fishbourne_sac_2004/downloads.cfm?CFID=43442&CFTOKEN=48E2A362-1C15-43A3-AF2D0A1928F0DB88&Google Scholar
Mauduit, C., and Lambert, P.Y. 2008: ‘Une découverte d'exception: le pot à fascinum de la Patte d'Oie à Gonesse (Val-d'Oise)’, in Rivet, L. and Saulnier, S. (eds), Productions, approvisionnements et usages de la vaisselle en Languedoc du Ier au IVe siècle apr. J.-C., Actes du congrès de Pézenas (25–28th May, 2006), Marseille, 617–25Google Scholar
Meggitt, J.J. 2013: ‘Did magic matter? The saliency of magic in the early Roman Empire’, Journal of Ancient History 1.2, 170–229CrossRefGoogle Scholar
Mees, B. 2009: Celtic Curses, WoodbridgeCrossRefGoogle Scholar
Nisard, M. 1864: Ciceron, Oeuvres complètes, vol. 4, ParisGoogle Scholar
Ogden, D. 1999: ‘Binding spell curse tablets and voodoo dolls in the Greek and Roman worlds’, in Ankarloo, B. and Clark, S. (eds), Witchcraft and Magic in Europe: Ancient Greece and Rome, Philadelphia, 1–90Google Scholar
Ogden, D. 2002: Magic, Witchcraft, and Ghosts in the Greek and Roman Worlds. A Sourcebook, New YorkGoogle Scholar
PGM = Preisendanz 1973–1974Google Scholar
Piranomonte, M. 2005: ‘La fontana sacra di Anna Perenna a Piazza Euclide tra religione e magia’, Revista Internacional de Investigación sobre Magia y Astrología Antiguas 5, 87–104Google Scholar
Piranomonte, M. 2010: ‘Religion and magic at Rome: the fountain of Anna Perenna’, in Gordon and Marco Simón 2010, 191–213CrossRefGoogle Scholar
Piranomonte, M. 2012: ‘Un contesto magico straordinario’, in Piranomonte and Marco Simón 2012, 161–74Google Scholar
Piranomonte, M., and Marco Simón, F. (eds) 2012: Contesti Magici. Papers from the International Conference held at the Palazzo Massimo, Rome (4–6th November 2009), RomeGoogle Scholar
Polakova, J., and Rapinesi, I.A. 2002: ‘I materiali magici’, in Piranomonte, M. (ed.), Il santuario della musica e il bosco sacro di Anna Perenna, Soprintendenza Archeologica, Milan, 38–52Google Scholar
Preisendanz, K. 1973–1974: Papyri Graecae Magicae (PGM): die griechischen Zauberpapyri, Stuttgart/Leipzig (original edition, 1928–1940, Leipzig/Berlin)Google Scholar
Rolfe, J.C. 1913: Suetonius, The Lives of the Caesars, Loeb Classical Library, Cambridge, MAGoogle Scholar
Rudling, D. 2008: ‘Roman-period temples, shrines and religion in Sussex’, in Rudling, D. (ed.), Ritual Landscapes of Roman South-East Britain, Oxford, 95–138Google Scholar
Santrot, J. 2007: ‘Lares et laraires en Gaule romaine: chapelle luxueuses et oratoires populaires’, in Baratte, F., Joly, M. and Béal, J.C. (eds), Autour du trésor de Mâcon: luxe et quotidien en Gaule romaine, Institut de recherche du Val de Saône-Mâconnais, 75–104Google Scholar
Scheid, J. 1981: ‘Le délit religieux dans la Rome tardo-républicaine’, in Le Droit religieux dans la cité antique, Collection de l'École Française de Rome 48, Rome, 117–71Google Scholar
Scheid, J. 2005: Quand faire, c'est croire. Les rites sacrificiels des Romains, ParisGoogle Scholar
Schilling, R. 1954: La Religion romaine de Vénus depuis les origines jusqu'au temps d'Auguste, ParisGoogle Scholar
Schmid, W.M. 1910: Das römische Kastell Abusina, MunichGoogle Scholar
SEG. Supplementum Epigraphicum Graecum, ed. Chaniotis, A., Corsten, T., Papazarkadas, N. and Tybout, R.A., Leiden/BostonGoogle Scholar
Seyrig, H. 1935: ‘Notes archéologiques’, Berytus 2, 42–50Google Scholar
Sindonino, S. 2011: ‘Reims, Rue Saint Symphorien’, Bilan scientifique 2011, ReimsGoogle Scholar
Sindonino, S. (ed.), forthcoming: Reims, Rue Saint Symphorien (Marne), unpub. Rapport Final d'Opération, Institut de recherches archéologiques préventives (INRAP)Google Scholar
Sindonino, S., Brunet, M., and Bailliot, M. forthcoming: ‘The magical deposit’, in Sindonino, forthcomingGoogle Scholar
Solin, H. 1968: ‘Eine neue Fluchtafel aus Ostia’, Commentationes Humanarum Litterarum 42.3, HelsinkiGoogle Scholar
Souville, G. 1957: ‘Figurines magiques trouvées dans le collecteur principal du decumanus maximus de Volubilis’, Bulletin d'Archéologie Marocaine 2 2, 180, pl. II, 3, 4Google Scholar
Spindler, K. 1984: ‘Anthropomorphe Terrakotten aus den römischen Lagerdörfern von Eining und Straubing’, Das archäologische Jahr in Bayern 1982/83, 113–15Google Scholar
Tomlin, R.S.O. 1988: ‘The curse tablets’, in B. Cunliffe, The Temple of Sulis-Minerva at Bath, Vol. 2, The Finds of the Sacred Spring, Oxford University Committee for Archaeology Monograph 16, Oxford, 4–277Google Scholar
Tomlin, R.S.O. 1993: ‘Votive objects: the inscribed lead tablets’, in Woodward, A. and Leach, P. (eds), The Uley Shrine: Excavations of a Ritual Complex on West Hill, Uley, Gloucestershire, 1977–1979, English Heritage Archaeological Report 17, Oxford, 113–30Google Scholar
Tomlin, R.S.O. 2008: ‘Paedagogium and Septizonium: two Roman lead tablets from Leicester’, Zeitschrift für Papyrologie und Epigraphik 167, 207–18Google Scholar
Trombetta, P.J. 1999: ‘La Guerre des Gaules, Phallus d'Ile-de-France et d'ailleurs’, in Religions, rites et cultes en Ile-de-France, Actes des Journées Archéologiques d'Ile-de-France (27–28th Novembre 1999, Institut d'Art et d'Archéologie, Paris), Saint-Denis, 83–115Google Scholar
Tupet, A.-M. 1976: La Magie dans la poésie latine, ParisGoogle Scholar
Turcan, R. 1992: Les Cultes orientaux dans le monde romain (original edition, 1989), ParisGoogle Scholar
Versnel, H. 1976: ‘Two types of Roman devotion’, Mnemosyne 29. 4, 365–410CrossRefGoogle Scholar
Versnel, H. 1987: ‘Les imprécations et le droit’, Revue Historique de Droit Français et Etranger 65, 5–22Google Scholar
Versnel, H. 1988: ‘A twisted Hermes. Another view of an enigmatic spell’, Zeitschrift für Papyrologie und Epigraphik 72, 287–92Google Scholar
Versnel, H. 1991a: ‘Some reflections on the relationship magic-religion’, Numen 38.2, 175–97CrossRefGoogle Scholar
Versnel, H.S. 1991b: ‘Beyond cursing, the appeal to justice in judicial prayers’, in Faraone and Obbink 1991, 60–106CrossRefGoogle Scholar
Versnel, H.S. 1998: ‘An essay on anatomical curses’, in Graf, F. (ed.), Ansichten griechischer Rituale. Symposium for Walter Burkert's Birthday (Castelen near Basel, 15–18th March 1996), Stuttgart/Leipzig, 217–68Google Scholar
Versnel, H.S. 2010: ‘Prayers for justice, East and West: new finds and publications since 1990’, in Gordon and Marco Simón 2010, 275–354CrossRefGoogle Scholar
Viret, J., Lambert, P.-Y., Stüber, K., Stifter, D., and Repanšek, L. 2013: ‘Le plomb de Chartres’, Études Celtiques 39, 123–92Google Scholar
Walke, N. 1965: Das römische Donaukastell Straubing-Sorviodurum, Bayerischen Akademie der Wissenschaften, Kommission zur archäologischen Erforschung des spätrömischen Raetien, BerlinGoogle Scholar
Wilburn, D. 2013: Materia Magica. The Archaeology of Magic in Roman Egypt, Cyprus, and Spain, Ann ArborCrossRefGoogle Scholar
Witteyer, M. 2005: ‘Curse tablets and voodoo dolls from Mainz’, Revista Internacional de Investigación sobre Magia y Astrología Antiguas 5, 105–23Google Scholar
Wünsch, R. 1898: Sethianische Verfluchungstafeln aus Rom, LeipzigGoogle Scholar
Wünsch, R. 1902: ‘Eine antike Rachpuppe’, Philologus 71, 26–31CrossRefGoogle Scholar
Figure 0

FIG. 1. (a) F1. Figurine from Mainz (after Witteyer2005, 111, fig. 4); (b) F3. Figurine from Mainz (Photo: Generaldirektion Kulturelles Erbe Rheinland-Pfalz, Landesarchäologie Mainz); (c) F4. Figurine from Eining (after Spindler1984, 113, fig. 96 right); (d) F5. Figurine from Eining (after Spindler1984, 113, fig. 96 left); (e) F6. Figurine from Straubing (after Spindler1984, 114, fig. 98); (f) F7. Figurine from Straubing (after Spindler1984, 114, fig. 97); (g) F8. Figurine from Petronell/Bad Deutsch-Altenburg (after Gassner2010, 224, fig. 3, Gassner et al.2011, pl. LXIX, fig. 18); (h) F9. Figurine from Fishbourne (after Bailliot and Symmons2012; photo: M. Bailliot); (i) F10. Figurine from Reims (after Bailliot2010, 110, fig. 25e; drawing: M. Brunet (INRAP)); (j) F10. Figurine from Reims and box made with cups (after Bailliot2010, 97, 103, 105, fig. 23; photo: INRAP); (k) F12. Figurine from Piriac-sur-Mer (after Hervé-Monteil2010, 52).