Episode 7

Curse Tablets

Hello, and welcome to the Poison Room. A podcast that’s something to do with texts… being dangerous. This week we are finally leaving the 18th century. In fact, we’re leaving it by quite a bit. Because today, we’re looking at some artefacts from the Greco-Roman world know as ‘curse tablets.’ As the name suggests, these things are tablets, and they contain curses.

In Greek, they’re called katadesmoi, in Latin, Defixiones. Both names draw on the idea that this is a type of ‘binding’ magic, and words referring to ‘binding’ are often used on the tablets themselves, which beseech supernatural powers to ‘bind’ or ‘restrain’ the target, and to confine them to a certain course of action – or inaction. Very rarely do they actually express a desire for the target to be killed.

Until a few decades ago, Classicists didn’t seem to have much interest in things like curse tablets. John Gager suggests that ‘one reason for this persistent neglect stems surely from the potentially harmful character of these small metal tablets – not so much from the real harm suffered by their ancient targets but the potential harm to the entrenched reputation of classical Greece and Rome, not to mention Judaism and Christianity, as bastions of pure philosophy and true religion.’ (1999: 3)

This old trope was used for centuries to support European colonial enterprises. Europeans were enlightened. European culture was better. And they traced this claim to superiority back to ancient Greece and Rome. Magic and superstition has no place in this enlightened classical past. Outside of far-right racist groups, that attitude is dying out, but it takes a while for academia to catch up and actually start producing quality research on previously neglected topics. We’re getting there with research on magic in antiquity, but it’s still a bit of a niche topic.

Despite past disinterest, curse tablets were definitely A Thing, and currently we have over 1,500 surviving examples of them. The earliest tablets we’ve found date to around the beginning of the 5th century BCE, and we’ve found them all over the place – in Roman Britain, North Africa, Egypt, Antioch, and more (Collins 2008: 65).

The subjects of curses vary across time and place. Curses are directed at commercial, legal, theatrical, athletic, and amatory rivals. The commercial, theatrical, and legal curses are mostly from around the classical and Hellenistic period – the earliest example for a commercial curse so far is from 450 BCE Sicily (Faraone 1991: 10). Curses targeting athletic performers are usually from the late Roman period, from the Secondnd century CE onward. In Athens, the curse tablets we’ve found that have targeted athletes – wrestlers and runners – are from the 3rd century CE, whilst the tablets targeting charioteers are from the 2nd century CE to the 3rd century CE and are basically exclusive to North Africa and Syria (Faraone 1991: 11, 13). Amatory curses that seek to separate people are found across the whole range of time curse tablets were being made, but those that seek to bring a couple together are a later development (Faraone 1991: 11).

With theatrical and athletic cases – and even legal cases – the tablets are concerned with a public event, in which many people have a vested interest in the result, not just the direct competitors. Sports fans want their athlete or team to win, citizens want political figures they dislike to do badly, so it’s not just the athletes or litigants involved in an upcoming confrontation that are making curse tablets – friends, spectators, political rivals, etc. are all getting in on the action, too.

Curse tablets to do with commerce or love, on the other hand, are generally personal – it’s far more likely that the issue is only of interest to the direct subjects of the scenario. In the case of curse tablets to do with love, there are two different types – those that sought to hamper competitors, and those that sought to encourage love in the target (though of course we have at least one exception that combines both, from the fifth century CE (Suppl. Mag. 45. Ogden p. 232).

Roger Tomlin, discussing a collection of curse tablets found in Bath, in England, and dated to around the second to fourth centuries CE, observed a further category unique to England: ‘prayers for justice’. Esther Eidinow describes the central features of these ‘prayers for justice’. They’re written by, or on behalf of, someone who has been wronged by others and predominantly relate to stolen goods. They tend to go beyond just asking the god for help and offering a justification for the petition, and frequently ‘dedicate the stolen item to the deity’s charge, or deliver the wrongdoer to the gods, or transfer the case itself to the god’s care.’ (Eidinow 2019: 351-2). Tomlin draws a stark distinction between the two, claiming that prayers for justice are not magical spells at all. He adds that the ‘prayers’ for justice do not contain many of the elements of curse tablets (Tomlin 1988: 62), and others have argued that they should be considered a separate category altogether (Versnel: 1991).

However, Eidinow also notes that, as is nearly always the case, distinctions that are easy to make in abstract are not so easy to make in reality. There’s disagreement over where exactly to draw the line, complicated by the existence of tablets that draw on both spells and prayers for justice. But one notable difference between prayers for justice and other curses is that the prayers for justice seek revenge for something in the past – to redress something deemed unfair, whilst the other curses are pre-emptive. Commercial, judicial, athletic, theatric, and amatory curses all focus on affecting the future actions and behaviours of someone, before they can harm the maker.

But what do these curse tablets – and prayers for justice – look like? Well, usually, they’re thin sheets of lead, inscribed with spells designed to invoke supernatural aid in harming a specific target. Other materials could be used, like papyrus, wax, and pottery sherds, amongst other things, but lead, lead alloys, and other metals were the favoured choice. Of course, metal tablets are far more likely to survive than papyrus or wax, so we can’t have a true sense of how much more popular lead was compared to these others. The sizes of the tablets can vary, with some being as small as 6 by 4 cm, and others closer to 24 by 30 cm. These sheets of lead were often rolled up on completion of the inscribing of the curse, and sometimes pierced with a nail, for added binding power. I’ll post pictures of some on Twitter, so you can see examples for yourselves.

Why lead? For several reasons, probably. The method of making a lead tablet is pretty simple (Tomlin 1988: 81). You pour hot lead into a mould, then roll, hammer, or scrape it flat and smooth. You can then cut it up into smaller sheets of the right size for your intended nefarious purposes. Lead was generally cheap and easily available, and was also a common medium for writing of any kind (Gager 1999: 4). But we also have at least one spell that instructs the user to just… nick the lead from a cold water pipe (PGM VII 396-404).

Finally, lead came to be seen has having the type of quality one frequently wished to inflict upon the target of the curse – excluding the amatory curses hoping to encourage love in someone. We have a tablet from the Athenian Agora that pleads that “just as these names are cold, so may the name of Alkidamos be cold” (Jordan, “Agora” p. 207). And there are others that ask the target to become as heavy as the lead (See Tomlin, p. 82 n 3). But these kinds of details don’t appear on the earliest tablets we have, so it seems that this was an idea that evolved once lead was already in use as a common material for the curse tablets (Gager 1999: 4).

Making the actual tablet is, of course, only part of the process. Next someone has to actually write on it. But who was doing the writing? We know from Plato that there were already professionals who offered their services in magic and curses in the third century BCE (Rep. 2.364 b-c), but based on the huge variation in the quality of the handwriting on the tablets we have, it seems that there were both professionals and amateur scribes making them, though Gager leans towards professionals being the dominant group, at least in the Roman period (1999: 5). Some tablets from that period appear to have been written in advance, leaving a space for the personal names to be inserted – though not always quite enough space – there are some tablets where the scribe has had to cram the name into the space. This is highly suggestive of someone offering a professional service (Gager 1999: 14), and expecting to do enough business that it’s worth having a supply of tablets written out in advance.

On the other hand, some of the tablets recovered from Bath are written in what Tomlin calls ‘pseudo-inscriptions’, where the text seems intended to look like actual writing, but is most likely an illiterate person’s attempt to mimic writing. This, coupled with the fact that very few of the Bath tablets appear to have the same handwriting, suggests that a professional might not always be available (Tomlin 1988: 247). Which also suggests that even without a professional available, people were committed enough to the idea of curse tablets being useful that they still made them anyway. But, as Tomlin also points out, it could mean that illiterate people didn’t want to cough up to get a professional to do it, or that perhaps Sulis Minerva – the local goddess – demanded petitioners write their own tablets. Whatever the case, it also suggests that not being able to write was not seen as a barrier to the efficacy of tablet, and that the goddess would still be able to understand it (Tomlin 1988: 247).

Writing on the tablets was, itself, not a difficult task. A metal stylus, often bronze, was used to make indents in the lead. But when you’re dealing with dark and powerful forces, getting the wording exactly right is often considered important, and that’s the type of thing that you might turn to a specialist for. Often we find tablets that share the same formula for the curse, which suggests that the scribe was working from some recipe they had recorded somewhere. You can think of it as a grimoire, if you don’t mind the anachronism. We have some of these formularies recorded in the Papyri Graecae Magicae or, in English, The Greek Magical Papyri, or, for short, the PGM. As you might expect from the title, the PGM is a collection of papyri from Greco-Roman Egypt, mainly from around the 2nd century BCE to the 5th century CE, containing spells and instructions for constructing spells, as well as hymns, and rituals.

So we’ve made the tablet. We’ve written, or paid someone else to write, on it. What does the spell, or curse, look like? What kind of things are they writing? Well, some of that’s going to depend on time and place. Earlier tablets from around the fifth to fourth centuries BCE tend to be the simplest, starting with tablets that give literally just the name of the intended victim. Some might include a verb, usually one meaning ‘bind’ and the name of a Deity (Gager 1999: 5).

As to the tablets that contain only the name of the intended victim, or victims, which make up the biggest group of Greek katadesmoi we have so far, Christopher Faraone has pointed to the similarities between these and names written on potsherds in connection with expulsion rituals and ostracism in 5th to 4th century democratic Athens. (Forsdyke 2005: 157). In fact, our word, ‘ostracism’ comes from the Greek word ὄστρακα (ostraka), which literally means ‘potsherd’, but also came to refer to the practice of writing someone’s name on a potsherd in order to vote for them to be ostracised.

In practice, ostracisms were relatively rare, but a vote was held every year on whether to have one, and if so, eligible participants would write the name of the person they wished to ostracise on a potsherd by way of a vote. It’s reasonable to assume that in these simplest tablets, the accompanying curse and the names of the deities invoked were spoken out loud. The shift to writing the details of the curse down on the tablet itself is unsurprising – writing has a permanency that the spoken or signed words lack. It gives a more concrete and permanent form to the idea. Having a crucifix with you wards off evil more effectively than just making the sign of the cross. These tablets that just list names become less popular overtime, and disappear from the record altogether in the first century CE (Faraone 1991: 5).

When it comes to the Greek tablets that also include the name of the deity, there’s a definite preferential order. According to Gager Hermes takes the top spot by a big margin followed up by Hekate, Kore and Persephone (Kore and Pesephone are the same, but Kore is the ‘daughter of Demeter’ version, and Persephone is the ‘queen of the underworld’ version). After that is Hades, then Gē or Gaia – the Earth, and then Demeter. Though according to Faraone, the top of the pops is Hermes, Ge, then Hekate and Persephone (Gager 1999: 12; Faraone 1991: 6). When it comes to the Latin tablets, the top spot goes to the spirits of deceased ancestors, known as Manes, then Jupiter in second, followed by Pluto, Nemesis, Mercury, and then a bunch of various water nymphs (Gager 1999: 12-13). In later tablets from Roman Egypt, the names of Egyptian deities – most commonly Thoth, Seth, and Osiris, also appear, and there are also names that are clearly from Jewish and Christian sources (Gager 1999: 14).

Why these gods? Primarily, they’re deities associated with death, or the underworld. Persephone and Hades are an obvious choice – they are, after all, powerful rulers of the underworld, and Hekate is another chthonic goddess – one associated with magic and crossroads. But why does Hermes take the top spot? The choice of Hermes might not be immediately obvious to you if you’re not particularly familiar with classics and ancient Greek myth. Hermes is a god whose remit includes merchants, travel, roads, and shepherds, amongst other things. He’s a bit of a trickster figure. ‘Travel’ doesn’t just include journeying around in the human world, it also includes the journey of the dead down to the underworld. In this role, Hermes acts as a psychopomp guiding the deceased down to Hades. In the Hymn to Demeter, it’s Hermes who goes to retrieve Persephone from the underworld and bring her back to her mother. So actually, Hermes is an ideal god if you want to get a message to the deceased, or the underworld.

As things start to get more complex, you start to find tablets that also address the dead alongside chthonic powers. Prime targets are those who died violent and untimely deaths. Because, of course, their spirits are restless, and angry, and hang around, unable to enter Hades (Johnston 2006: 17).

Other elements start to appear as time goes by and tablets start to get more complicated. Sometimes the name of the target of the curse would be written with the letters scrambled. Sometimes it would be written backwards. Sometimes the whole thing would be written backwards. These deviations from normal writing are details embedded in the text that are symbolic of the kind of thing you want the curse to accomplish – to stop, to confuse, to tangle up, to reverse progress. They’re also the kind of things that, if you’re a professional, make what you’re doing seem more complicated, dangerous, and mysterious. It adds an aura of mystique. Magic should not appear normal or mundane. The more complicated a spell, the more a lay person will worry about getting it wrong or making a mistake, the more likely they are to pay someone else to do it for them, to make sure that it’s right and will do what they want it to do.

Other textual elements that can be found on curse tablets are palindromic words, and letters arranged in geometrical shapes. The technical terms for these shapes formed from letters is carmina figurata. A common form is to start on the first line with a full word, and on subsequent lines remove one letter, until you end up on the final line with only one letter remaining – or vice versa. Or sometimes both. Both of these types of writing emphasise that the text is not normal: it doesn’t follow the formatting of text humans use to write to each other, because it’s not intended to act as communication between two living, normal, mortals. As David Frankfurter points out, it also suggests that the visual representation of the letters themselves are thought of as having special significance, beyond the sounds they represent. (2019: 637).

Later Roman tablets contain an increasing amount of voces mysticae, or voces magicae - words which don’t appear to belong to any language, and whose meaning is not obvious. But the fact that we can’t understand it doesn’t mean it wasn’t thought to have meaning. Gager explains that, when it comes to addressing spirits and underworld figures, ‘it would be entirely inappropriate to use one’s native, human, ordinary language’ (Gager 1999: 10). It’s probable that the introduction of some of the voces magicae was due to the influence of Egyptian traditions (Eidinow 2019: 278).

We might not be able to understand the words, and, indeed, the person writing them might not have understood them either (Gager 1999: 10), but they did think that this language would be intelligible to the entities they were invoking. Humans use human languages. Supernatural beings have their own language. And we, being humans, can’t understand it. Gager, however, also notes that many of these voces mysticae have been discovered to be “real” words from other languages, such as Aramaic, Hebrew, Persian, and various forms of ancient Egyptian (Gager 1999: 10). These voces mysticae might also be combined in a formula – recurring again and again in the same order and shape. In recipes, the instructions for these recurrent formulas in spells is often abbreviated, so instead of the whole formula being written out, you might find an instruction to use ‘the abc-formula’ (Gager 1999: 9).

Here’s an example of what your curse tablet might look like so far, from the PGM. The weird words are the voces mysticae:

I adjure you, Evangelos, by Anubis, Hermes, and the remaining powers of the underworld, to bring and bind Sarapias, to whom Helen gave birth, to Heraeis, to whom Thermoutharin gave birth, now, now, quickly, quickly. Draw her by her soul and heart to Sarapias, to whom Helen gave birth with her own womb, MAEI OTE ELBOSATOK ALAOUBETO OEIO AEN. Bring and bind the soul and heart of Sarapias, to whom Helen gave birth, to Heraeis, to whom Thermoutharin gave birth with her womb, now, now, quickly, quickly. (PGM XXXII 1-19)

This curse is a love curse of attraction, seeking to bind together Sarapias and Heraeis. If you think that those both sound like women’s names, that would be because… they’re both women’s names. Yes. This is an amatory binding spell between two women. Women loving women is not a new thing. You can see the petition to various gods, the identification of the targets of the spell, and the voces mysticae.

Here’s another one:

I deposit with you this binding spell, gods of the underworld, Pluto and Kore yesemmeigadon and Koure Persephone Ereschigal and Adonis, who is also barbaritha, and underworld Hermes-Thoth phokensepseu earektathou misonktaich and mighty Anubis pseriphtha, who holds the keys to the (gates) of Hades, and (with you,) chthonic spirits, gods and goddesses, who suffered an untimely death, boys and maidens, year after year, month after month, day after day, night after night, hour after hour. I adjure you, all spirits in this place, to assist the corpse-demon. (Rouse yourself for me, corpse-demon,) whoever you are, whether male or female, and go into every place, into every quarter, into every house, and bind Kopria, whom her mother Taesis bore, the hair of whose head you have, for Ailourion, whom his mother named Kopria bore, that she may not submit to vaginal or anal intercourse, nor gratify another youth or another man except Ailourion only… (P.Mich 757)

This one’s from the University of Michigan’s papyri collection, and is dated around the third to fourth century CE. It’s another amatory curse, and it addresses both gods and the spirits of the dead. It’s part of a larger text from a papyrus found in Egypt, and casts the central actor as the ‘corpse-demon’ or nekydaimon. The prominence of the corpse-demon is particularly prevalent in the later spells found in Egypt, which might well be to do with the Egyptian traditions of rituals to do with writing letters to the dead, where the dead person was named, and asked to perform a specified action (Frankfurter 1997: 117-35).

Having so far pointed out how the words and language of curse tablets differ from everyday writing, it should also be noted that they do share some of their features with other types of writing – personifications, the use of metaphor, repetitions, threats, and prayers, etc. place them within a class of writing, which also includes general religious prayers, legal styles, and even the oral poetry traditions (Gager 1999: 13-14). Curses might borrow the language and phraseology of the courts (Frankfurter 2019: 628). Legal language has a sense of authority, of power – winning or losing a legal case has visible, tangible, real-world consequences. Using that kind of authoritative language makes sense. To a modern audience, the fact that curses draw on magic power sets it apart from these other things, but to an ancient audience, this is simply a variation on the power behind poetry, prayers, and legal language, and the boundaries between all those things are blurred.

Note that the binding spell from the PGM starts with the invocation of the gods, then petitions for the two women, giving both their names and their mother’s names, to be drawn together ‘now, now, quickly, quickly’, then come the voces mysticae, then the petition repeats, word for word, until the ‘now, now, quickly, quickly’. There’s a lot of duplication going on, which really emphasises the qualities of the text as formal, ritualistic, or even legalistic. Just think of the terms of conditions we totally all read every time we sign up to something. How repetitious the language there can sound, or how carefully laws or legal complaints might be structured to make sure there’s no doubt at all over who the person specified is or what the targeted action is. The listing of the names of the mothers of the women help make sure the correct Sarapias and Heraeis are being identified. Gotta make sure it’s Sarapias, daughter of Helen, the spell targets, and not just any woman who happens to be called Sarapias.

Another element that appears in later tablets – from the 2nd century CE onwards is the use of charatēres. These are symbols that appear in a wide range of spells, not just curse tablets (Gager 1999: 11). The power of magic symbols is still a common trope today. Just think of shows like Supernatural, where such symbols are used to summon, bind, or banish various paranormal entities. I’m not going to go into depth (here, at least) on the power of symbols, but it’s something most of us have some sort of understanding of – we understand that a single symbol can stand in for a complex ideology – from political parties, to hate symbols to brand identification. They’re powerful tools, so finding them in magic, where you’re trying to invoke powerful beings, is wholly unsurprising. I’ll post an example of what these * charatēres* looked like on twitter.

Drawings also appear on many curse tablets, usually representing either the human target or the supernatural entity being named and invoked, and it’s not uncommon to find effigies used to support curse tablets. In the National Library in France there’s a 4th century BCE curse tablet in the shape of a person, from Karystos in Euboea (you-bee-a). It kinda’ looks like the gingerbread man. Andrew Wilburn notes that it seems that it was first made as a 3D effigy, and then pancaked (Wilburn 2012: 488). It’s not hard to imagine that there’s a ritualistic significance to doing it that way round, rather than cutting out the figure from a pre-flattened sheet of lead.

In the PGM several of the texts containing formulae for spells instruct the practitioner on what to draw to accompany the texts (Wilburn 2019: 489-492). I’ll post photos of an example on twitter, and it’s well worth a look. It’s PGM (XXXVI 231-55), and shows a delightfully child-like drawing of a figure holding the head of another. Now, ancient drawings can be pretty hard to interpret, but here’s part of Wilburn’s interpretation of the drawing: ‘the vignette depicts a rooster-headed, cuirassed divinity, the anguipede, who holds a sword in one hand and a possibly decapitated head in the other’ (Wilburn 2012: 491). The anguipede is a figure commonly found on magic devices – gems, papyri, etc. Frequently it has serpents for legs, but this isn’t the case in our anguipede. We don’t actually know who specifically the image of the anguipede is supposed to represent. Various theories have been suggested, but there’s no definitive consensus (Nagy 2002), though the name Abraxos comes up frequently. Another drawing of the anguipede appears earlier on the same papyrus, this time holding what appears to be a whip in one hand, and a full representation of a tiny human in the other (XXXVI 86-100), and in total the anguipede, holding various things, appears six times in the spells on this papyrus alone.

Okay. So. You’ve got your nice, smooth, lead tablet. You’ve filled it with words, strange languages, invocations to various gods, symbols, and drawings, all helping to indicate who you want to invoke and what you want to happen to your victim. Whilst you’ve been doing that you might have read out part of a spell to accompany it, or you might recite a spell during the next part. And you might have been collecting or making other items to go along with the tablet. Such as effigies – little doll-like representations of humans. They were usually made of lead or wax, and nearly all have the arms of the figure in a position tied behind it’s back – literally binding it (Gager 1999: 15). They also might be pierced by nails or needles. So by now you’re probably imagining something that’s definitely bad and harmful magic, and that the pins are designed to hurt or harm that part of the person.

Except… These figures also accompany things such as love spells, so as weird or ominous as they might look, the spell they’re attached to is not necessarily a physically harmful curse (though trying to make someone fall in love with you against their will is undeniably totally immoral and gross). And as for the nails? Usually they help indicate the parts of the victim’s body that the curse is intended to affect, but that’s not the same as intended to inflict pain. For instance, Tongues and hands were frequently targeted in in the case of judicial curses – to prevent the victim from successfully presenting their case in court. For athletics competitions, it might be arms and feet. So it’s not so much ‘hurt this body part’ as ‘inhibit this body part’, or, in the case of amatory spells, ‘fill this body part with desire’.

The coolest example of this is a set of effigies from the Kerameikos in Athens that dates from around 400 BCE: four lead figurines, each sealed in its own little lead coffin, with the names of the intended victims inscribed in the inside of the coffins and on three of the figurines themselves. I’ll post a photo of these on twitter, too.

What’s next step? You put it somewhere significant, of course! Frequently, that means ‘bury it’. Common sites for burying curse tablets are graveyards and crossroads. Shocking, I know. Magic at crossroads and in graveyards? How very unusual. The tablets themselves also give locations such as rivers, and wells, among others (PGM VII 451-2). If the tablet is intended to hamper a rival in athletics, then the stadium itself is also an option. If it’s a love spell then you might aim to put it in or near the home of the target, or at a public bath. Graveyards seem to be the most common, which makes sense if your curse tablet is addressing the spirits of the dead, though obviously, archaeologists are more likely to excavate a graveyard than a crossroad, and tablets dropped in the sea or a river are going to be pretty hard to find.

Now, obviously curse tablets fit the theme of this podcast because they’re texts. And they’re considered dangerous, and capable of inflicting harm. But, as Gager points out, the creation of the tablet is only part of a process:

As the recipes from the PGM clearly demonstrate, the total process could be quite complex. It required invocations, purifications, fumigations, prayers, instruments, rituals, and more… In short, the client’s attempt to prepare a defixio, from the initial decision to its actual commission, must be located in a series of actions, a total flow of events rather than a single isolated act. It is a serious mistake to focus attention solely on the innocuous piece of corroded lead and not to perceive other actions that accompanied its commission. (1999: 20).

The written curse tablet did not function in isolation. It might be the most tangible part of the spell, but that doesn’t mean it’s the most important. We need to be careful that we don’t place undue emphasis on its importance as a single component, just because it’s the part that has survived in our archaeological records.

Not all of the elements I’ve mentioned are present in every curse. In the PGM there are spells that only mention the creation of the curse tablet, alongside spells that don’t require anything written at all, and spells that have complex instructions involving both the creation of the curse tablets, down to details such as what ink and writing implement to use, as well as instructions on where and how to place the tablet once made, and spells that might need to be said at some point in the process.

As an example, let’s turn back to our old friend, PGM XXXVI.

Take a lead lamella and inscribe with a bronze stylus the following names and the figure, and after smearing it with blood from a bat, roll up the lamella in the usual fashion. / Cut open a frog and put it into its stomach. 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 rising of the sun. (XXXVI 231-55).

I’ll post a photo of the things you’re supposed to write and draw on twitter. Note how many different elements there are, here. It’s not enough to just write the required words. You need to use a bronze stylus, smear the tablet with blood from a bat, cut open a frog and put the tablet inside, and sew the frog back up with Anubian thread and a bronze needle. You need to hang it from a reed using hair from the tip of the tail of a black ox, and you need to put it in the right place.

All of these details are equally important. Forgetting the bat blood, or using hairs from something other than the tail of a black ox, or placing it at the west of your property, rather than the east, will all interfere with the efficacy of the spell just as much as making a mistake in writing out the words themselves. And of course, acquiring all the items necessary – a bat, a frog, a reed, a hair, Anubian thread, a bronze stylus, and needle, as well as the tablet, are all part of the process too.

So we’ve covered the basics of the process of creating tablets. Let’s pause for a second to consider a question which might not have occurred to a sceptically-inclined modern audience: did they work?

The simple answer is ‘yes’. People were using these things because they thought they worked. The same way theists today believe in the efficacy of prayer. But what about instances where the tablets clearly failed to work? If you’re already committed to a system of belief, especially one as ritualised and complicated as production of a curse tablet could be, then failures can easily be explained as being caused by incorrect performance of the ritual, or it being overpowered by a counter-curse. And confirmation biases would also be at play. If you believe curse tablets will work, then you will place more weight on evidence that supports that believe than on the evidence that contradicts it. That’s just standard human behaviour.

Sure some people, like Plato, who speaks scornfully of those who use them (Laws 933b 2–3), clearly thought they were nonsense. But Plato is a philosopher. We can’t use his personal views as indicative of the general consensus or opinion on the matter. Indeed, the very fact that he mocks people who do use this stuff is indicative of the opposite being true: many people took it seriously. You don’t complain about people doing something unless, y’know, they’re actually doing that thing.

Gager makes an interesting point with regard to this question: if we ask it from a starting place of believing ourselves that they don’t work, then our answers to this question only focus on explanations of why such practices were thought to work when they clearly don’t. But if we accept that practitioners genuinely did believe they worked, then our answers take the form of explaining in what sense they might have been thought to work. Should we perhaps interpret the texts in a more metaphorical light? I.e. did people consider the curse effective even if it didn’t produce the specific result they sought?

There has been a lot of research demonstrating a positive impact of expressive letter writing, from aiding with the management of chronic pain (Graham et al. 2008), to improved quality of sleep after writing about negative life experiences (Mosher & Danoff-Burg 2006). And anyone who’s ever felt helpless will be familiar with the desire to just want to do something. Curse tablets provide a thing you can do.

Of course, this doesn’t prove that curse tablets were thought to function because they had a positive impact on the person making or commissioning the curse tablets in terms of being able to deal with what had happened to them, or might happen to them – remember, most curse tablets were pre-emptive, after all – and it would be incredibly rash to suggest this was the line of thinking held at the time, given that there’s no evidence to support it. But it certainly shows that we can’t rule out the possibility that these effects were part of what helped people believe they did work.

On top of that, we know that preparing generally helps performance. And if curse tablets are a common part of preparation for a legal case, then it’s easy to see both how the enacting of a preparatory ritual might aid performance, and, if you’re performing better than your opponent, that you might attribute their inferior performance to your curse tablet. And if making a curse tablet helps you feel better, or helps you perform better and beat the opponent you cursed, then does it matter if the curse literally came true?

Okay, so we know how to create a curse tablet, and even have some idea as to why they were thought to work. Now let’s look at some other questions: who was using magic? How common was it? And what were the different attitudes toward it?

The profound impact of Christianity on Western culture means that today, many Westerners, when thinking of magic, think of something socially taboo – possibly even Satanic. It is something that is done by only a few people, and done in private, secretly, and not spoken of. Not so, in the ancient world. The broad range of uses curse tablets were employed for indicates that people from all walks of life made use of them. All the evidence we have suggests that curse tablets, and magic more generally, were a popular thing.

Now, I know what your next question is: what can you do to protect yourself against these curses? Well, you could try and take legal recourse, or you could try… more curses!

Of course if you’re used to the idea of pre-emptively cursing people to achieve desired outcomes, then the idea of pre-emptively protecting yourself from other people’s curses is gonna’ occur to you. Here’s an example of an amulet found in a grave, from Amisus in Pontus, Asia Minor.

I am the great one who is sitting in heaven, the wandering hollow of the cosmos ARESNONEOPHRIS, the safe name MIARSAU as the true daimon BARICHAA KMÊPHI who is the ruler of the kingdoms of the gods ABRIAÔTH ALARPHÔTHO SÊTH. Never let evil appear. Drive away, drive away the curse from Rouphina; and if someone does me an injustice, revert (the curse) back to him. Nor let poison harm me. King of kings ABRIAÔN TO ORTHIARÊ. I am the one ruling the place in Moses’ name.

The date for this amulet is uncertain, but it’s probably from the first century CE. Now when I say ‘amulet’, we’re not talking about what you might immediately think of with the term – this one is a tablet of silver, rolled up and placed in a small bronze case, and was found in a grave (Gager 1999: 225-6). But amulets could also be worn by the living, tied around necks, or arms, or legs. They could be embroidered bands or just pieces of string, rings, or engraved stones, or rolled up bits of papyrus or metal. Amulets have a broad range of uses – to ward of illnesses, disasters, harm, danger, or curses, but also, sometimes, for the same kind of purpose as amatory curses, or to grant success. (Gager 1999: 220).

What about legal recourse? Is it available to you? Or, if you’re wanting to make a curse tablet yourself, do you need to worry about someone trying to take you to court?

There were no specific laws against katadesmoi in the law codes from classical Greece (Gager 1999: 23), which might lead us to conclude that the societies in question didn’t consider it as a concern or something that needed legislating against, but, as Charles Phillips III points out, legislation is often created by a small, elite group of society, whose concerns and attitudes cannot be assumed to mirror that of the majority of the society, and may be subject to the influence of politics (Phillips III 1991: 261). Part of the problem with figuring out the legal status of curse tablets is that the laws tend not to mention them specifically. Instead they tend to use words that on the one hand could be used to refer to specific parts of magic, but were also used more broadly to indicate magic as a whole. For example, in 479 BCE, the Ionian Greeks at Teos declared the production and use of pharmaka dēlēteria illegal. Pharmaka could be used to refer to potions specifically, but here, the term seems to mean ‘magic’ in general and thus could easily be understood to include curse tablets (Gager 1999: 23).

Plato lets us know that he thinks there should be a law against binding spells, but he also thinks that they don’t work and are just silly, superstitious nonsense. So he wants to ban them because he thinks they’re a bad influence, and emphatically not because he thinks they work (Laws XI 933a). But that’s Plato, and all this tells us is what Plato himself thought.

Now, when it comes to later periods, and the Romans, we have more evidence of what people thought about curse tablets. Clearly, they were still in popular use. They’re more frequently being used to target athletic performers, charioteers (in North Africa and Syria), and even gladiators. Tablets seeking to make the victim fall in love with someone are more common.

The first evidence we have for laws against magic in Roman law is from the Twelve Tables, from around the 5th century BCE. Now, the original text of the Twelve Tables has not survived, which means we can only access it through the details given by other, much later writers. And as we all know, interpretations of laws can change over time. For instance, when Pliny, writing in the mid 1st century CE talks about Table eight, which contains three parts that might be considered as aimed at outlawing magic, he interprets the first part as being aimed at those who cast evil spells (VIII 1). But earlier, Cicero, in the first half of the first century BCE, interpreted the same law as meaning poetry aimed at ruining someone’s reputation (quoted by Augustine de Civit Dei 2, 9). The second two parts are concerned with ‘bewitching crops’ and magically luring them from others’ fields to your own (VIII 8a-b). Now, even Pliny thinks this idea is absurd, but he does think magic is real, and mentions quite a few incantations that people were using to treat various medical ailments. When a philosopher called Apuleius was defending himself at trial from accusations of using harmful ‘incantations’, in around 155 CE, he cites this law about crops as providing the basis for a general law against magic.

Apuleius was also charged with using poison and love potions, a charge grounded in the Lex Cornelia de sicariis et veneficis from 82-81 BCE. The statute was originally issued by the dictator Sulla and was originally intended to deal with terrorism and restore public order (Kippenberg 1997: 147). Once again the statute itself has not survived, but it’s discussed by the ever useful Cicero. Cicero frames it as a law against having anything to do with the process of manufacturing or using fatal poisons. The word used is venenum, and like pharmaka can have a narrower meaning of ‘poison’, or a broader meaning that includes using magical means. This ambiguity allowed for lawyers to expand the law to cover spells and rituals (Kippenberg 1997: 148).

There were also whole bunch of trials going on in the first century CE among political elite, who were vying and scheming for power, as political elites are wont to do. At periods of high political turmoil, such as under the rule of Tiberius, accusations of magic were flung about, and usually paired with accusations of treason, or other things. So amongst the 1%, magic is being used as a convenient charge that can be used to bolster other accusations against a rival who you want to get rid of. But there’s an element of that present in the trial of Apuleius, too. See, he upset some people by marrying Pudentilla, a widow. Doing so was financially bad for members of her family who weren’t her new husband, and on top of that, Pudentilla was like, ten years older than him, and that’s just weird, right? So you can read part of the motivation being to get rid of someone for financial gain, and part of it as a general attack on someone for breaking accepted social norms (Kippenberg 1997: 142). Which is not to say that they didn’t also think that Apuleius had legit used magic to both make Pudentilla fall in love with her, and then to kill her son when she tried to tell him she was under a love spell.

All in all, it’s a complicated picture, and there’s a development over time towards interpreting old laws to encompass the outlawing of any sorts of magic, and different uses of the laws in different social groups. On top of that, there’s also a shift from a focus on punishing the use of magic because of the harm specific instances caused, to an idea that magic itself is intrinsically immoral (Gager 1999: 24; Rivers 2002).

There’s just a tiiiiiiiiny possibility that this had something to do with the rise of Christianity. A document of legal rules compiled around 300 CE is very anti-magic. People doing magic get burned alive. The same kind of laws were issued by Constantine, the first Christian emperor. What can I say, Christianity loves a good witch-burning (Kippenberg 1997: 149).

The physical evidence for curse tablets peters out in the 7th century CE and disappears from the 8th century on. And Gager suggests that this was ultimately due to the fact that curse tablets were simply too pagan to be incorporated into Christianity, as so many other pagan practices and traditions were. (1999: 29-30).

That’s in for curse tablets for now. Undoubtedly in the future I’ll come back to some of the texts mentioned, and maybe focus on some of them, and there will inevitably be a lot more episodes that deal with magic versus Christianity.

Thanks for listening. If you like the show, please subscribe on whatever podcatcher you’re using. Rate and review the show, especially on iTunes.

If you have questions, comments, feedback, want to suggest a topic, etc. You can find the podcast on twitter @poisonroompod, or send an email to: poisonroompodcast@gmail.com

Alternatively, find some lead pipes and steal a flat bit of lead, paint the lead with the blood of a dodo, write down your thoughts on it, making sure you do so with a stylus made of bronze. Sew it up inside a dead animal and place it at the end of your property and hope the neighbours don’t get too suspicious about the smell. I won’t read it, but it’ll be fun.

Transcripts of this episode are available at poisonroom.com, where you can also see the references and bibliography. Again, if the sources are publicly available, they’re linked to.

You have been listening to The Poison Room, a podcast that definitely did not just accidentally summon a corpse-demon.


Betz, H. D. (1986) The Greek Magical Papyri in Translation, Including the Demotic Spells. Vol. 1: Texts, University of Chicago Press.

Collins, D. (2008) Magic in the Ancient Geek World, Blackwell Publishing.

Eidinow, E. (2019) ‘Binding Spells on Tablets and Papyri’ in Guide to the Study of Ancient Magic (Ed. Frankfurter, D.) Brill, 351-387.

Faraone, C. A. (1991) ‘The Agonistic Context of Early Greek Binding Spells’ in Faraone, C. A, & Obbink, D. (Eds) Magika Hiera: Ancient Greek Magic and Religion, Oxford University Press. 3-32.

Forsdyke, S. (2005) *Exile, Ostracism and Democracy, Princeton University Press.

Frankfurter, D. (1997) ‘Ritual Expertise in Roman Egypt’ in Envisioning Magic Schafer, P. & Kippenberg, H. (Eds), Brill. 115–35.

--- (2019) ‘The Magic of Writing in Mediterranean Antiquity’ Guide to the Study of Ancient Magic (Ed. Frankfurter, D.), Brill. 626-658

Gager, J. G. (1999) Curse Tablets and Binding Spells from the Ancient World, Oxford University Press.

Graham, J. E., Lobel, M., Glass, P., & Lokshina, I. (2008) ‘Effects of Written Anger Expression in Chronic Pain Patients: Making Meaning From Pain’ Journal of Behavioural Medicine Vol. 31: 201-212.

Johnston, S. I. (2006) ‘Magic and the Dead’ in Greek Magic: Ancient, Medieval, and Modern Petropoulos, J. (Ed.) 14-20.

Kippenberg, H. G. (1997) ‘Magic in Roman Civil Discourse: Why Rituals Could be Illegal’ in Envisioning Magic Schäfer, P. & Kippenberg, H. G. (Eds), Brill. 137-163.

Martinez, D. G. (1991) Michigan Papyri XVI, A Greek Love Charm from Egypt (P.Mich. 757) Scholars’ Press.

Mosher, C. E., & Danoff-Burg, S. (2006) ‘Health Effects of Expressive Letter Writing’ Journal of Social and Clinical Psychology Vol. 25: 1122-1139.

Nagy, Á. M. (2002) ‘Figuring out the Anguipede (‘snake-legged god’) and his relation to Judaism’ Journal of Roman Archaeology Vol. 15: 159-172.

Ogden (2002) Magic, Witchcraft, and Ghosts in the Greek and Roman Worlds: A Source Book, Oxford University Press.

Rivers, J. B. (2003) ‘Magic in Roman Law: The Reconstruction of a Crime’ Classical Antiquity Vol. 22: 313-339.

Suetonius (1889) (trans. Thomson, A.) The Lives of the Twelve Caesars, Gebbie & Co.

Tomlin, R. S. O. (1988) ‘The Curse Tablets,’ in The Temple of Sulis Minerva at Bath Vol. 2. The Finds from the Sacred Spring B. Cunliffe, B. (Ed.), Oxford University Committee for Archaeology.

Versnel, H. (1991) ‘Beyond Cursing: the Appeal to Justice in Judicial Prayers’ in Faraone, C. A. & Obbink, D. (Eds) Magika Hiera: Ancient Greek Magic and Religion, Oxford University Press. 60-106.

Wilburn, A. (2012) Materia Magica: The Archaeology of Magic in Roman Egypt, Cyprus, and Spain, University of Michigan Press.

Wilburn, A. (2019) ‘Figurines, Images, and Representations Used in Ritual Practices’ in Guide to the Study of Ancient Magic (Ed. Frankfurter, D.) Brill, 456-506.

Wray, D. (2002) ‘Astrology in Ancient Rome: Poetry, Prophecy and Power’ Fathom Archive, University of Chicago.