Textual Magic: Charms and Written Amulets in Medieval England

By Katherine Storm Hindley

An expansive consideration of charms as a deeply integrated aspect of the English Middle Ages.
 
Katherine Storm Hindley explores words at their most powerful: words that people expected would physically change the world. Medieval Europeans often resorted to the use of spoken or written charms to ensure health or fend off danger. Hindley draws on an unprecedented archive of more than a thousand such charms from medieval England—more than twice the number gathered, transcribed, and edited in previous studies and including many texts still unknown to specialists on this topic. Focusing on charms from 1100 to 1350 CE as well as previously unstudied texts in Latin, French, and English, Hindley addresses important questions of how people thought about language, belief, and power. She describes seven hundred years of dynamic, shifting cultural landscapes, where multiple languages, alphabets, and modes of transmission gained and lost their protective and healing power. Where previous scholarship has bemoaned a lack of continuity in the English charms, Hindley finds surprising links between languages and eras, all without losing sight of the extraordinary variety of the medieval charm tradition: a continuous, deeply rooted part of the English Middle Ages.

• Language: English
• Publisher: University of Chicago Press
• Release date: Aug 16, 2023
• ISBN: 9780226825342

Book preview

Cover Page for Textual Magic

Textual Magic

Textual Magic

Charms and Written Amulets in Medieval England

Katherine Storm Hindley

The University of Chicago Press   CHICAGO AND LONDON

PUBLICATION OF THIS BOOK HAS BEEN AIDED BY A GRANT FROM THE BEVINGTON FUND.

The University of Chicago Press, Chicago 60637

The University of Chicago Press, Ltd., London

© 2023 by The University of Chicago

All rights reserved. No part of this book may be used or reproduced in any manner whatsoever without written permission, except in the case of brief quotations in critical articles and reviews. For more information, contact the University of Chicago Press, 1427 East 60th Street, Chicago, IL 60637.

Published 2023

Printed in the United States of America

32 31 30 29 28 27 26 25 24 23     1 2 3 4 5

ISBN-13: 978-0-226-82533-5 (cloth)

ISBN-13: 978-0-226-82534-2 (e-book)

DOI: https://doi.org/10.7208/chicago/9780226825342.001.0001

Library of Congress Cataloging-in-Publication Data

Names: Hindley, Katherine2w3 Storm, author.

Title: Textual magic : charms and written amulets in medieval England / Katherine Storm Hindley.

Other titles: Charms and written amulets in medieval England

Description: Chicago ; London : The University of Chicago Press, 2023. | Includes bibliographical references and index.

Identifiers: LCCN 2022047656 | ISBN 9780226825335 (cloth) | ISBN 9780226825342 (ebook)

Subjects: LCSH: Charms—England—History—To 1500. | Magic—England—History—To 1500. | English literature—Middle English, 1100–1500—History and criticism. | English literature—Old English, ca. 450–1100—History and criticism. | Latin literature, Medieval and modern—England—History and criticism. | Anglo-Norman literature—History and criticism.

Classification: LCC GR600 .H56 2023 | DDC 203/.32094202—dc23/eng/20221122

LC record available at https://lccn.loc.gov/2022047656

This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

Contents

List of Figures

List of Boxes

Note on Translation and Transcription

Abbreviations

INTRODUCTION  Reading, Writing, and Charming

CHAPTER 1  The Powers of Charm-Words and Relics

CHAPTER 2  Before 1100: Textual Magic in Pre-Conquest England

CHAPTER 3  1100 to 1350: Charm Language and the Boundaries of Text

CHAPTER 4  1350 to 1500: A Fayre Charme on Englysh

CONCLUSION  The Changing Power of Words

Acknowledgments

Manuscripts Cited

Works Cited

Index

Footnotes

Figures

0.1  Depictions of the history of medicine (fifteenth century)

0.2  Apollo performing healing (fifteenth century)

1.1  Manuscript birth girdle (dorse) (late fifteenth century)

1.2  Amuletic image of the cross (fifteenth century)

1.3  Beholde and se (late fifteenth century)

1.4  Beholde and se (late fifteenth century) [digitally enhanced]

2.1  Abracadabra amulet (thirteenth century)

2.2  Gold lamella (third or fourth century)

2.3  The Franks Casket (early eighth century)

3.1  Ciphered medical text (early twelfth century)

3.2  Protective characters (late eleventh century)

3.3  Abbreviations in a charm for nosebleeds (early twelfth century)

3.4  Magical figure incorporating the word tetragrammaton (thirteenth century)

3.5  Magical seal (thirteenth century)

3.6  Charm for wishes (fourteenth century)

4.1  Magical seal for protection (fifteenth century)

4.2  Magical seal for protection against storms (fifteenth century)

4.3  Transliterated Greek in a charm against cramp (fifteenth century)

Boxes

0.1  Selected Charms Claiming to Function through the Grace of God

0.2  Selected Charms Incorporating Spoken, Written, and Liturgical Components

0.3  A Charm for Wounds

1.1  Veronica Charms for Bleeding

1.2  A Charm against Fevers

1.3  Selected Charms for Childbirth with Dissolved Text

1.4  A Charm for the Falling Evil

1.5  Fever Charms Written on Communion Wafers

1.6  A Fever Charm Written on Sage Leaves

1.7  Childbirth Charms Written on Butter or Cheese

1.8  Nichasius Charms

1.9  Apollonia Charms

2.1  Pre-Conquest Charms Including Runic Characters

2.2  Pre-Conquest Charms Using Names as Efficacious Texts

2.3  Pre-Conquest Parallels between Past Success and Present Expectation

2.4  In Principio Erat Verbum in Pre-Conquest Charms

2.5  Pre-Conquest Charms Using the Pater Noster

2.6  A Charm for Elf-Hiccup

3.1  French Charms

3.2  Charms with English Instructions

3.3  Charms with French Instructions

3.4  Variations of the Tigath/Acre Charm

4.1  Charms Giving Instructions in Multiple Languages

4.2  English as the Sole Language of Power in Charms

4.3  English Instructions, French Efficacious Words (and Vice Versa)

4.4  Textual Charms in the Vernacular

Note on Translation and Transcription

Unless otherwise specified, translations are my own. I have placed footnotes after quoting the original language when the translation is mine and after quoting the translated text when citing another translation.

When quoting published transcriptions from manuscripts, I have used the editorial conventions of my source. When I have transcribed text myself, I have expanded abbreviations with the supplied letters in italics and raised letters lowered. Letters presented in square brackets are missing from the manuscript but have been supplied from context, as, for example, when space has been left for the future addition of rubricated initials. Ellipses in square brackets represent damaged or illegible text, while ellipses without square brackets are used when I am deliberately omitting words from the quoted passage. Bold text represents manuscript rubrication, while text that is underlined or crossed out in the manuscript source is similarly underlined or crossed out in my transcription. I have normalized word separation where it would otherwise make reading difficult. I have not marked line breaks or insertion marks for scribal corrections.

Abbreviations

Introduction

Reading, Writing, and Charming

In 1382, a London court sentenced Roger Clerk to ride bareback through the city to the sound of trumpets and pipes, carrying around his neck two urine flasks, a whetstone, and a fraudulent textual amulet.¹ Roger’s crime was falsely impersonating a physician; the textual amulet he carried was the one he had tried on his patient Johanna. Both the expectation that written words might have healing power and the specific details of Roger’s trial shed light on the complex relationships between literacy, medical practice, and the efficacious power of words in medieval England.

The crux of Roger’s trial as it was described in the records lay in attitudes toward the written word. The supposed amulet was merely an old parchment . . . a leaf of some book rolled up in cloth of gold, yet Johanna and her husband had initially believed that the material text and page might have the power to heal her of her unspecified maladies.² During the trial, Johanna’s husband presented the offending object to the court as evidence. When Roger appeared,

dicta cedula ei monstratur per Curiam et quisitum est ab eo ad quod dicta cedula valet qui dicit in illa scribitur bonum carmen pro febribus Vlterius quesitum est ab eo per curiam que sunt verba predicti carminis sui qui dicit aima christi sanctifica me corpus christi salua me isanguis christi nebria me cum bonus christas cum laua me et inspecta cedula predicta nullum eorundem verborum in ea scripta fuerunt.³

the said parchment was shown to him by the Court, and he was asked what the virtue of such a piece of parchment was; whereupon, he said that upon it was written a good charm for fevers. Upon being further asked by the Court what were the words of this charm of his, he said;—Anima Christi, sanctifica me; corpus Christi, salve me; in isanguis Christi, nebria me; cum bonus Christus tu, lave me. And the parchment being then examined, not one of those words was found written thereon.⁴

Unable to even get the words of the suspect charm right, Roger was found guilty.

This book is a study of words at their most powerful: words that, according to the beliefs of their users, could be deployed to physically change the world, providing healing to the sick or protection to the vulnerable. In Roger’s trial, we see two key points of tension surrounding such efficacious words: first, the Latin—at least as it has been recorded—was garbled, and second, the amulet did not contain the words that Roger claimed.⁵ Both of these relate, in different ways, to the fear that people might misuse or mangle powerful words they did not understand or wrongly assume that powerless words had power. As we shall see over the course of this book, concerns about whether or not efficacious words could properly be understood and concerns about who could read or access them were persistent. Medieval authorities repeatedly condemned the use of unknown words, whether written or spoken. Roger was found guilty not only because he had lied but because he was in no way a literate man and so could not understand what his parchment said—a fact the brief record of his trial reports not once but twice.⁶ Although educated people could and did use written charms, Roger was not a fit practitioner.⁷ As his trial demonstrates, some charms required literacy.

In using textual magic in the title of this book, I seek to disrupt the assumptions about uneducated practitioners that might otherwise surround the word charm. First, in emphasizing text, I wish to highlight that these charm-remedies make use of a technical skill—writing—that was not available to all. Second, in emphasizing magic, I aim to draw attention both to the practical effects that charms sought to produce and to the ambiguity of the power that lay behind their words. Broadly speaking, medieval commentators considered magic to originate either from demons or from occult properties inherent in certain natural objects, often thought to be caused by the rays of the stars or planets.⁸ While many charms invoked nonmagical sources of power, such as God or the saints, others did not make the origin of their power clear and were therefore open to accusations of being demonic in nature. It is often not evident whether a person performing a charm would have perceived him- or herself as practicing medicine, as calling on the hidden natural properties of the words themselves, or as carrying out an act of religious faith. For this reason, I use the neutral term practitioner throughout this book, since it can refer to either a medical practitioner or a practitioner of magic.

In the study that follows, I explore conceptions of the power of textual magic in medieval England from the pre-Conquest period to the end of the fifteenth century. This was a period of tremendous change in which the inhabitants of England altered their writing systems and their languages and during which both literacy rates and access to the written word soared. To examine the importance of writing as a medium through which patients could interact with words, I place examples of efficacious text—in which the charm functions by means of the written word—alongside protective and healing uses of spoken words. As we shall see, people in need invoked power in both ways throughout the Middle Ages, although the specific uses of spoken and written charms diverged.

Previous scholarship on charms in England has tended to focus on the pre-Conquest period, which has been well studied.⁹ Most recently, Ciaran Arthur’s Charms, Liturgies, and Secret Rites in Early Medieval England has argued that the very concept of charms in this period should be abandoned—an idea I address in chapter 2.¹⁰ Despite the attention paid to early charms, the broader English charm tradition has been neglected relative to the traditions of many other European countries. In his 2005 study English Verbal Charms, Jonathan Roper notes that there is no monograph dedicated to English charms . . . nor any halfway comprehensive catalogue.¹¹ Roper’s work, based on a database of five hundred verbal charms spanning more than a thousand years, took the first important step toward filling that gap. However, there is still no catalog of English charms and still no monograph focused on later medieval English charms, or on charms in England during the Middle Ages as a whole.¹² My book aims to remedy this lack. The texts I have collected over the course of researching this book, amounting to more than eleven hundred medieval charm-copies, many of them previously unpublished, will be combined with further charm-copies and made available in an accompanying database.¹³

Although my focus is on England, the medieval culture of charm use reached across Europe and beyond. Some charm-types were widespread in many medieval regions and languages. These include the Three Good Brothers charm against wounds, the Peperit charm for childbirth, and various versions of the Heavenly Letter for protection, all of which will be explained and discussed in the pages that follow.¹⁴ Other types were particularly common in certain language groups: T. M. Smallwood notes that several charm-motifs, including the Flum Jordan charm against bleeding, are shared across the countries of the North Sea and are found in a total of many hundreds of copies or recorded oral occurrences in the languages of northern Europe, notably German and English, and in next to none of the Romanic languages.¹⁵ There was also local variation within related language groups in terms of which charms were recorded and the conditions to which charms were applied. For instance, W. L. Braekman’s collection of medieval Dutch charms lists several incantations to treat burns, a condition not treated by any of the charms in my collection from medieval England.¹⁶ And yet, in the postmedieval period, the burn charm Out Fire, in Frost was one of the most common narrative charms in England but was not adopted into other languages.¹⁷ As these brief examples demonstrate, charm practices were common across Europe but varied by time, place, and language. Some types of charms existed in many countries; others were restricted to particular language families, and still others to certain geographical regions. While charm culture was widespread, its specifics could be, and often were, quite local. This book therefore examines the English manifestation of a practice with much broader reach.

Charms, Literacy, and Elite and Popular Culture

Given the focus on literacy and understanding in Roger Clerk’s trial, it would be easy to see objections to charms as reflecting learned, literate attempts to moderate popular religious practice. However, the situation was never that simple. Charms, like other elements of medieval religious belief, were subject to the complex patterns of influence that connected popular beliefs and folklore with medieval scholarly writing. This dynamic was brought to the fore in the 1980s with the spirited debate between Jacques Le Goff and Aron Gurevich on the origins of purgatory. Le Goff’s arguments emphasize scholastic developments that gave shape to medieval thinkers’ previously amorphous ideas about penance after death. He links the developments to new socioeconomic structures and to Latin theologians’ deliberate opposition to the Greek church.¹⁸ In Le Goff’s model, then, scholarly ideas shape and direct popular beliefs. Gurevich, by contrast, argues that while the precise configuration of purgatory may have been indistinct, its function was clearly established in popular culture before it was formulated by scholars. In Gurevich’s view, the beliefs and concerns of ordinary people influenced the ideas of those who preached and ministered to them, driving university scholars to articulate popular theories in academic form.¹⁹

Although Le Goff and Gurevich disagree on the direction of cultural influence, both stress the significant overlap and interplay between scholarly and popular beliefs. Charms, too, reveal such interplay. As I demonstrate in chapter 1, written charms—which were not discussed much by theologians—were understood to function in ways parallel to healing relics, which were discussed in depth. Whether this indicates that medieval scholars built their theories on a set of shared assumptions about the transmission of healing power, or whether it suggests that medieval Christians generally applied church teachings more widely than theologians intended, scholarly and popular ideas were intertwined.

The argument that charms demonstrate interplay between scholarly and popular beliefs does little, however, to clarify what popular culture actually was. Various definitions could apply: Gurevich, for example, suggests that popular culture might refer to the culture of the lower classes of society, or to the culture of the illiterati as opposed to the educated.²⁰ Neither of these divisions quite reflects the reality of medieval charms.²¹ Modern assumptions about charms—verbal formulas used for practical purposes, including healing—often associate them with oral performance, with women, and with the uneducated. This was also true in the Middle Ages: the fourteenth-century author of the Fasciculus morum, for instance, assumed that the user of made-up charms would be some wretched old woman.²² Yet this picture is inaccurate. Medieval charms, as others have rightly argued, were used by men and women from the whole social spectrum and sometimes relied on the willing involvement of priests.²³ Spoken charms certainly circulated orally—a mode of transmission which unfortunately leaves no trace—but they were also sometimes recorded by university-trained physicians and surgeons.²⁴ Charms that require their texts to be copied assumed at least basic literacy on the part of their practitioners. Their use crosses the boundaries that terms such as elite and popular suggest.

The early modernist Peter Burke’s model of great and little cultures helps to explain elite participation in popular practices such as charming. Burke holds that the great tradition of society, transmitted at schools and universities, was restricted to the educated elite, while the little tradition, transmitted informally, was accessible to all.²⁵ These ideas have been applied compellingly to the Middle Ages, as in Richard Firth Green’s study of medieval fairy beliefs.²⁶ But their application to the Middle Ages has also been challenged. Carl Watkins, for example, argues that Burke’s model cannot function for the medieval period because the institutions of the Church did not set out to preserve the articles of religious belief in the same way that grammar schools and universities sustained Burke’s elite culture.²⁷ Rather than keeping theological learning exclusive, the church deliberately found ways to bridge the gap between scholarly teachings and local practice. Watkins accordingly proposes an adaptation of Burke’s theory, arguing that there were "two traditions in medieval religious culture but all members of medieval society participated to a greater or lesser degree in both of them."²⁸ Rather than compartmentalization, Watkins sees a spectrum on which official and unofficial beliefs mingled.²⁹

Although the intentions of medieval institutions may have differed from those of early modern ones, and cultural boundaries may therefore have been less sharp, it is undeniably true—as Watkins acknowledges—that some forms of learning never reached the parishes.³⁰ In practice, then, there were aspects of educated culture in which the uneducated did not participate. However, it is also true that Burke’s model allows insufficient room for the great tradition to transmit religious ideas to the little one.³¹ I therefore envisage a situation somewhere between Burke’s and Watkins’s ideas.

In the specific case of charms, practical access to the remedies also complicates the idea of great and little traditions, as there is a gulf between efficacious words that must be spoken and efficacious words that must be written. Spoken charms could be repeated by anyone who could remember their words, whether or not they also understood their meaning. When charms required texts to be written down, they would have to be performed by someone with some level of literacy, and therefore with some level of access to the great tradition. However, people in medieval England used many languages, including English, Latin, French, Old Norse, Cornish, Dutch, and Hebrew.³² Latin literacy was generally more restricted than vernacular literacy, but the ability to read, write, or understand any one of these languages did not necessarily imply the ability to read, write, or understand others. Nor did the ability to read a language necessarily imply the ability to write it.³³ Some people may therefore have been capable of reading and understanding a written charm without being able to perform it themselves. In the discussions that follow, when I use the word literate with no further qualification to refer to the user of a charm, I intend to refer to someone sufficiently literate in the language used by that charm to perform it, including writing it down if that was required. In other words, I am interested in whether a particular charm could have been performed by many people or only by the comparatively well educated.

Although my work is concerned with issues of orality and literacy, my focus differs from that of major works such as Brian Stock’s Implications of Literacy and Walter J. Ong’s Orality and Literacy: The Technologizing of the Word.³⁴ Whereas both of these works are concerned with the effect of increasing literacy on the organization of society and language, the present book aims to explore the effect of rising literacy on the belief that the written word could hold or transmit physical power. The texts that Stock and Ong study were influential because they were read; the texts I am interested in were also influential yet were often not intended to be read, in the traditional sense of the term, at all.

As we have seen in the case of Roger Clerk’s Johanna, the text of a charm was not necessarily used by the person who created it. This means that even charms whose performance required literacy could have been used by illiterate patients. Indeed, most charm instructions suggest that the person who performed the charm was separate from the patient, whether the words were to be spoken or written. There are only rare cases in which the performer and the patient appear to have been the same. One of the clearest examples is a thirteenth-century childbirth charm that exhorts Deleuere me of mine childe, a clear indication that the woman in labor would say the charm herself.³⁵ The original composer of a charm, the charm practitioners, and the patients on whom the charm was used may all have had different levels of literacy and different forms of participation in the great and little traditions of their societies.

Some aspects of charm culture must stem from, or be influenced by, the great tradition of the educated and literate. One obvious example is the use of Greek, even in corrupt form, in periods when Greek learning would have been restricted to a small number of highly educated readers.³⁶ Sometimes the charm practitioner had to copy texts using the Greek alphabet, while in other cases he or she was instructed to say Greek phrases aloud.³⁷ Users could doubtless have repeated or copied the words without understanding their meaning, but the composers of the charms, at least, must have had some access to the great tradition. The charms therefore reflect some awareness of a literary culture that was inaccessible to all but the highly educated, as well as the desire to harness its prestige. Similarly, as I argue in chapter 3, some written charms have no semantic meaning but borrow the appearance of heavily abbreviated texts that might be read by the educated. In these cases, the great tradition also influences the little one, because it provides the visual and verbal cues through which the charms convey their power.

During the roughly seven centuries covered by this book, literacy in England increased substantially. The average person’s perception of the power of written words presumably changed along with it. For someone who cannot read and has no understanding of how reading works, there is no clear connection between the letters on the page and the sounds or meanings those letters convey. Someone with a little more skill in reading might be able to decipher some or all of the letters but not the meaning of the text. A fluent reader, meanwhile, would be able to read the words aloud, but might or might not understand the language in which they were written. Each of these imagined readers understands more clearly than the last the relationship between written letters and the meanings they communicate.

The chronological organization of the central section of this book aims to illuminate how the use of written words as conduits or providers of healing power changed as increasing numbers of people came to understand this relationship. Broadly speaking, two major shifts occurred. First, there appears to have been a substantial increase in the proportion of written charms recorded in the post-Conquest period, possibly due to new cultural influences. Second, late medieval charm instructions are more concerned with secrecy, perhaps reflecting a larger audience of literate users who might misuse the texts of charms or disseminate texts that charmers preferred to keep to themselves. I explore these developments in detail in chapters 3 and 4.

Charms: Objections, Definitions, and Distinctions

Although charms were sometimes criticized or condemned, the practice of charming was also seen as central to medieval medical practice. It was not merely a last resort for people without access to other forms of medical care or for people whose diseases could not be cured using other treatments available at the time. Indeed, charms could be the first resort for patients, as in one remedy that promises "to stanche blood withouten charme whan a master vayne ys for koruen and wol nat gladly stanche with charme."³⁸

The major role charms played in medieval medicine is vividly apparent in the illustrated history of medicine in one fifteenth-century copy of the works of the English surgeon John Arderne (d. in or after 1377).³⁹ Ink drawings, arranged in tiers, introduce the reader to key medical advances: Aesculapius, the Graeco-Roman god of medicine, hands a bowl of medicine to a patient, teaches men to gather medicinal roots and herbs, holds a scale to explain the importance of measurements and quantities, and makes complex medicines with a pestle and mortar. The series ends with Hippocrates and Galen, depicted on either side of an ornate table covered with pots and jars. The caption indicates that they discovered "certeyne quantitez in reseyuyng, perhaps an error for the qualitez" of humoral medicine (fig. 0.1). The verso of the leaf on which these images appear has been reproduced several times.⁴⁰ However, it is the very first image in the series, on the recto, that is the most intriguing for the study of charms.

Fig. 0.1 British Library, Sloane MS 6, fol. 175v. The image captions read, in the first tier, ". . .  resonable gouernance of law of lywyng, in the second, Esculapius helyd men with fermcis [i.e., pharmacies] medicinez and Aschepius taught to gedere rotes herbez flourrez frotez, and in the third, Asclepius schewed mesures quantites weghtez wases, Asclepius techeþ to mak pulueres confeccionis electuaries, and Ypocras galen scheweþ certeyne quantitez in reseyuyng." Photograph: © The British Library Board.

At the bottom of the recto of the manuscript leaf is a depiction of Apollo, whom the highly influential Isidore of Seville called the author and discoverer of the art of medicine.⁴¹ Isidore associated Apollo’s school of medicine with remedies and charms, but here the herbal remedies are ignored and only charms are depicted (fig. 0.2).⁴² Apollo sits on an elaborate throne, gesturing with his right hand toward a book held open in his left. In front of him, three men, eyes gently closed, stand at a strange diagonal that seems to represent a state of trance. The caption reads, "Apolo helyd men with chermez inschantementez. A different hand has labeled the image further, writing apolo across Apollo’s chest. On the pages of Apollo’s book, this hand has written, disapprovingly, þe booke deuill," presumably meaning that the book itself is devilish.

Fig. 0.2 British Library, Sloane MS 6, fol. 175r. Photograph: © The British Library Board.

This series of images emphasizes that charms and enchantments form the very origins of medical practice, depicting them as the key innovation of medicine’s founder. For this artist, medicine begins with magic. It also appears, in this series, that medicine begins with a book. Although the second image in the series also shows a practitioner holding a book, that figure’s book rests in his lap while his eyes and gesturing hand engage his patient.⁴³ In this first image, on the other hand, Apollo’s book is apparently in active use, held between himself and his patients. The book was also the focus of the later reader’s particular objections: only the book, presumably along with the texts the reader imagined it to contain, was labeled as devilish. This drawing therefore encapsulates themes that recur throughout my study: the important position of charms within medical practice, their reliance on written or spoken words, and the religious tensions that surrounded their use.

Although writers often opposed the use of charms on religious grounds, the definition of charming was not always stable. The fourteenth-century Lollard Apology for Lollard Doctrines, for example, condemns people who gather herbs while saying "ani charme but þe pater noster, or þe crede."⁴⁴ The author considers the Pater Noster and Creed acceptable for use in gathering rituals, but he nevertheless appears to place them under the category of charme when they are used for that purpose. Elsewhere within the same text, he argues that charms necessarily act against the will of God, stating that charms

are þat are brout in bi fendis curst, and bi stering of fendis, aȝen þe bidding of God, and also be mannis vanite and foly, wiþ out ground of God Almiȝti, and in wilk men trystun of help wiþ outun him, and oftun aȝen as ȝeþun and vnfeiþful don; þus we callid charmers þo þat wil bi þer curst haue a þing þow it plece not God. And þis schewiþ what is a charme, weþer it be charme maad or writun, or þe wirking of þe charmar.⁴⁵

What identifies a charm here is not its external presentation, but the invisible mechanism of its power and its use in opposition to God’s plan. Given the fact that commentators’ views of what a charm is depend on circumstance and perception as well as on the practice itself, the question of how to define a charm requires some elaboration.

The English word charm and the French charme derive from the Latin carmen, meaning song or incantation.⁴⁶ Although the Latin meaning emphasizes oral performance, by the end of the Middle Ages the terms charm, charme, and carmen could all be used to refer to words that had to be written down in order to transmit their power. Spoken charms might also be called an incantation or adjuration, while written charms might be described as a writ or breve. The word charm, therefore, could be used as an umbrella term encompassing both spoken and written words, while other terms might be used to distinguish between efficacious text and speech.

For the purposes of this study, I consider charms to be written texts or spoken words presented as having a specific practical outcome associated with their enactment, of which the most common forms were recitation (performance in speech) and writing (words being physically inscribed on some receptive surface). I consider texts with a primarily or purely devotional purpose to be prayers. However, as I discuss below, the dividing line between charms and prayers is by no means straightforward. Even in cases where texts clearly seem to function as charms, promising specific practical outcomes, they frequently refer to divine power. For example, a small number of texts that I classify as charms include a clause stating that they will only be effective with the grace of God (box 0.1). This brings the charm securely into a Christian worldview in which God holds ultimate power, but it does not alter the link between a performance involving specific words and an expected physical outcome. The overall situation is further complicated by the fact that medieval Christians believed God to have imbued natural things such as plants, stones, and words with hidden properties. These properties included healing and protective powers that could be activated without direct divine intervention. Charms might therefore be seen to work simultaneously through God’s power and through the natural properties with which he imbued his creation. An example of this can be seen in one method of prognostication that claims to function þorw þe grace of god and þe verueyne.⁴⁷

Box 0.1 Selected Charms Claiming to Function through the Grace of God

British Library, Sloane MS 1314, fol. 37r–v (ca. 1350–1400)

Forto wyte ȝyf amon schal lyf þat ys seke Take .v. croppes of verueyne in þi ryht honde ley hem in lyft honde sey ouer hem fyue Pater nosters in þe worschyp of þe .v. woundes of + cryst sey þus. I. coniur ȝou croppes fyue in þe vertu of þe .v. woundes of + crist þat he suffred on þe rode tre forto bring mannes soule out of thraldam þat þe man þat is seke telle me þe sothe þorouhe þe vertu of god of ȝou. wheþer he schal lyf or dye of þat seknesse blesse hym .v. tymes take hem þen in þi ryht hande ageyn go to þe seke take his ryht hande in þin aske hym howe he fares whil þe herbes arn betwene þat he wot not howe he hopes of hym self he schal þorouhe þe grace of god þe vertu of þe verueyne telle þe sothe sikerly.

Cambridge, Magdalene College, Pepys MS 1661, pp. 82–83 (ca. 1400–1450)

¶ And of þe praptike of almagest an experiment y preued thow we ne lyȝe no fey þer to as vs semeþ more þat it be whycchecraft þanne wele dom of ony redy man and one þer of is with þe which þer haue y be many y heled tak þis experiment to þe festre preued. ¶ Take a staf and whenne þu by gynnest to enchauntye sey Pater noster to þe ende and In nomine patris et filii et spiritus sancti amen. and ȝif þu be wiþ inne house throwe þe staf out at þe wyndowe and ȝif þu be in a courte þrowe þe staf ouere þe ȝate bote loke þat þu þrowe it noȝt al ouer þanne go to þe place þer egermoyne groweþ and or euere þu dygeþ it vp þat þere by leue noþing in þe erthe of þe rote turne to þe est and whenne þu drawest vp þe rote hast done sey pater noster a noon to set libera nos and whenne þu drawest vp þe rote with þe leues þanne take þre partyes or ii. parties of þe ouerere and on of þe neþere whanne þu takest þe neþer partye sey þis word poynt ȝemon ȝeberam ȝaual abnari paraclitus and whenne þu takest þe ouerere partie sey pecor ȝemon wiþ alle þat oþer wordes out take poynt and þese iij. parties on newe linene cloþ euery partye be hym selue on þre knottis of þe cloþ and euery partie on his owne knotte fram euery oþer and bynde þe cloþ with summe þred seynge on on euery knotte Pater noster and wiþ inne ix dayes aȝeyn sey al þe experiment and so with grace of god þe pacient schal be delyuered.

Durham University Library, MS Cosin V.V.13, fols. 24v–25r (ca. 1475–1500)

for a woman that travils with chylde wryte thys ley it on hyr wombe Maria peperit christum Anna mariam Elizabeth Johannem O Alma remigium O And yeve hir to drynke diteyne or ysoppe with warme water she shall be deliuer Anon with þe grace of god.

Speech and writing, the two methods of accessing the perceived power of words, present different problems for researchers. All the spoken charms I discuss are in some sense also written charms in that they necessarily have survived to the present day in written form. My evidence is therefore the product of literate people recording practices that also circulated orally. As the spoken performance of a charm is a temporary and transient phenomenon, written texts are the only records of medieval spoken charms that we can possibly have. However, just as a cake