ARTICLE

The Acoustemology of the Witch:
Hearsay, Sound Recording, and the Zaccheo Tapes

Delia Casadei and Marina Romani

Sound Stage Screen, Vol. 3, Issue 2 (Fall 2023), pp. 5–45, ISSN 2784-8949. This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. © 2024 Delia Casadei and Marina Romani. DOI: https://doi.org/10.54103/sss22961.

Fig. 1. One of the witches’ crossroads in Zaccheo. Photo: Marina Romani (2023).

We can begin with an episode recovered from a taped interview recorded in August 1966. The speaker is a farm hand, Francesco, who recalls his encounter with witches in the countryside surrounding Villa Zaccheo (commonly known as simply Zaccheo), a village in the Teramo province of Abruzzo, some thirty years before, at a crossroads in the pitch black of night. [1] Though we listen to Francesco on a digitized tape, in a sound archive in the province of Novara, Italy, he is being recorded at the age of 58 in August 1966 by historian Cesare Bermani (in whose sound archives the tape is stored). [2] At the time of recording Francesco is in his home in Zaccheo, likely at the kitchen table, surrounded by Bermani as well as five women who giggle knowingly as he speaks: his daughters Lillina (31 years old), Lucia (27), Natalina (26), and his wife Concetta (52); with them is also Mariafelice Forti (28), Bermani’s collaborator, interpreter from dialect into Italian, and then-wife. The event narrated by Francesco dates back to roughly 1935, when Lillina was a sickly infant and suspected to be tormented by witches at night. [3] We, the present-day listeners, join Francesco’s audience in 1966 as he leads them back in time to 1935, to the crossroads and into the darkness where the meeting with the witches is about to occur. He tells the story expertly, to the wry laughter of his companions at the table:

Francesco presents an archetype encountered in the oral history of witchcraft in this area: it involves standing watch at a particular place and time so as to convince the witches to stop bothering a family member, often an infant who has fallen ill. A striking architecture of eye and ear often emerges from these stories, revealing a common theme of seeing and being seen, hearing and being heard. It involves an act of mutual recognition between the witch and her witness that happens, however, at a place and time when one cannot see but only hear: the open countryside at midnight.

To this day, the village of Zaccheo is minimally lit, and the area where Francesco went to see the witches in 1935 would have been in complete darkness. Lamps were not part of the protocol that would have allowed one to see the witches (unlike pitchforks, children’s clothes, and carpet beaters, which were used as protective objects). [6] There was an injunction to keep total silence when standing at the crossroads at night: as Francesco tells us, the witch is recognized there and then, but can be verbally addressed only later, in the daytime. The aural and visual restrictions of the encounter are counterbalanced by the witness’ subsequent compulsion to narrate it as many times as possible, and by the role of active listening from women—audibly marked by their laughter in the above recording. The encounter with the witch is a carefully staged missed connection: you may not speak to a witch as she is witching (or else, Francesco knows, she will kill you), and seeing her and recognizing her while she roams in the dark is highly unlikely. Directly seeing and even confrontation, where witches are concerned, are difficult tasks. Yet it is precisely this blocking of frontal, visual, and verbal encounter that produces an alternative, and intensely sonorous, sensoriality.

Another term for the alternative sensoriality swirling around the invisible and unapproachable figure of the witch might be acoustemology—Steven Feld’s famous portmanteau joining “acoustic” and “epistemology.” [7] Feld famously argues that an acoustemological analysis tracks the ways in which, in certain conditions, sound becomes a privileged carrier of knowledge in a society. In the analysis that follows, we argue that the tapes recorded by Bermani illuminate the means by which, in Zaccheo, the realm of hearing becomes a privileged means of knowledge in relation to witchcraft. Our analysis offers insights into acoustemology as it takes on new aspects at a contemporary and, we hope, potentially fruitful conjunction of sound studies and oral history, where witchcraft may cast light (or, better still, darkness) on broad themes.

By way of demonstrating what acoustemology can bring to an oral history of witchcraft, we might observe that the most interesting part of the recorded interview transcribed and translated above is what happens after Francesco has finished his story. The interview, initially a question-reply format led by Bermani, turns into a choral undertaking, with Francesco’s daughters, his wife, and Mariafelice eclipsing the interviewer’s role. Bermani has asked the women, one by one—in a strange echoing of Francesco’s “una ad una, una ad una…”—if they believe in witches, if they know how to defend themselves against them, what they learned about the evil eye, werewolves, saints, and miracles. Each of the women has openly denied believing in witches or knowing anything about the topic, and they have done so in Italian, albeit a dialect-inflected Italian: given that Bermani is from Novara, in north-western Italy, the choice of language shows their willingness to be recorded and understood by outsiders. Yet, once these initial answers have been given, true speech can begin: a debate in dialect on whether witches will disappear as newer generations stop hearing about them. This exchange is aurally opaque, aimed away from the interviewer’s foreign ear and microphone, but it is much more urgent and controversial for the interviewees. Hearsay evidence and opinions are offered liberally: variations of the expressions dece (they say/it is said) and so sendète (I’ve heard) punctuate the tangle of voices and the rhythm of the conversation:

Few admit to believing, but all feed passionately upon the hearsay through which the beliefs are created; and in feeding on it, they feed it in turn, even encouraging the notion that new witches are being made and roam lands nearby. There is also an important sense here—one which we will go on to qualify in political and historical terms—that the figure of the witch doesn’t entirely overlap with the imagination of the witch inherited from Counter Reformation Inquisition tribunals: the witch as an object of violent persecution. In the world summoned by these tapes, by contrast, individual witches can be managed, subdued, and turned back into friends. They are not persecuted or killed, nor is their eradication deemed desirable. [9] Indeed, the witch has a communitarian function. When Francesco and the others gleefully point out that witches might keep being born elsewhere, the implication is generative, as if to speak and hear of a witch is already to bring her forth into the world.

Sendè, the dialectal inflection of the verb sentire (to hear)—so present in this single episode—is commonplace in people’s accounts of witches in all of the tapes we heard in Bermani’s archive. As with all romance languages (and as famously theorized by Gilles Deleuze) the verb sentire means both “to hear” and “to feel,” “to sense.” [10] This slippage between hearing and states of being is heightened in the dialectal form: sendè atypically takes the auxiliary “to be” rather than “to have” (as is the case with sentire in Italian and most romance languages). [11] This means that, in Zaccheo, “I have heard” becomes so sendète—which can be more literally retranslated as “I am heard.” Even more interestingly, in the world of the Zaccheo tapes on witchcraft, sendè in its transitive form can also mean to “hear about” witches; in other words, it can designate hearsay, indirect witnessing, and the relayed experience of another. [12] Therefore, the sentence li sdraje li so sendète (which can be literally translated as “witches, I am heard them”) can be more idiomatically translated in two different (yet not easily separable) ways: as “witches, I have heard them” and “witches, I have heard about them.”

In its polyvalent meaning, sendè opens up a series of questions about the acoustemology of witchcraft in Bermani’s Zaccheo tapes, resulting from a dynamic interplay between the denial of knowledge/belief in witches and the consequent feeding of hearsay. As mentioned above, the interviewees’ initial denial of knowledge/belief in witches is almost a ritual for generating a subsequent, passionate group conversation on witches, often made up of multiple, overlapping voices and laughter. Another deeply acoustemological question pertains to the historical and political importance of relaying, through hearsay, events where sight spectacularly fails and hearing strains to fill the gap. As we will go on to argue, in exploring the relationship of hearsay, hearing, and the temporary loss of sight that signals the presence of the witch, we are also urged to reconsider the definition and function of “acousmatic” listening (broadly defined as the listening to sounds whose source cannot be identified). Lastly, we consider this acoustemology as a product of the particular ways in which interlocutors respond to the microphone and tape recorder.

The Zaccheo Tapes

In this essay, we focus on the remarkable collection of tapes that Cesare Bermani recorded in Zaccheo—tapes that are, in large part, dedicated to interviews about local practices of witchcraft, witches, and ways of warding off evil magic. Marina Romani is a native of Zaccheo and fluent in the local dialect; she is familiar with the surviving interviewees as well as surviving family members of those who are deceased. Our return to this set of tapes is an unusual methodological choice, given that the tapes have already been the subject of a remarkable book by Bermani himself: Volare al sabba. Una ricerca sulla stregoneria popolare(Flying to the Sabbath: A Research on Popular Witchcraft), published in 2008. [13] Indeed, our work has been developed in consultation with Bermani, who has welcomed us into his personal archives and allowed us to listen to the tapes. In listening to them, we are building on Bermani’s research and extending it towards a systematic consideration of the role of sound and listening in oral interviews on witchcraft. We are, in other words, carrying out an intellectually ecological mission: instead of creating new fieldwork to investigate the matter of witchcraft (something we plan to do at a later stage), we are drawing out an historical acoustemology by relying on recordings made by a scholar of a previous generation. This is in itself unusual and represents a model for how to recycle and redeploy ethnographic recordings: [14] one that accounts for the historical and political conditions of the recordings themselves, as well as the interviewees’ relationships with one another, with the place, and with the recording apparatus.

Some important information is now in order to explain the various layers of this “second-hand” acoustemology. As we mentioned above, Villa Zaccheo (henceforth Zaccheo) is a rural village in the province of Teramo, Abruzzo, with a population of around 300 people at the time of Bermani’s fieldwork. Abruzzo is a region that, while geographically part of central Italy, has belonged—politically, culturally, and linguistically—to the south, that is, to the Kingdom of Naples, until Italy’s unification in 1860. Bermani wound up in Zaccheo in 1959 in order to visit the family of his fiancée (and soon-to-be wife) Mariafelice Forti, whom he met at the University of Milan as a fellow student of philosophy. The Forti family had lived in the village for generations, where they were one of the major landowners, and were part of a close-knit community. Bermani would visit Zaccheo twice every year from 1959 until 1976, becoming a site of long-term anthropological fieldwork. Following an initial spurt of interest in local traditional songs, [15] Bermani’s focus progressively shifted to documenting the belief, common among Zaccheo’s inhabitants, that witches caused illness and mischief in the area. Between 1965 and 1976, Bermani officially narrowed his focus to witchcraft and conducted extensive interviews with 54 people—around one quarter of the town’s inhabitants at the time—as well as numerous additional interviews with people living in neighboring villages and cities, coming from different professions, generations, and educational backgrounds. [16] Such is the richness and extension of the materials gathered by Bermani—more than 90 hours of recorded interviews—that Volare al sabba could only make partial use of them. Indeed, Bermani makes qualitative, rather than quantitative use of the tapes: in the book, he leaves an extraordinary amount of room to his narrators, and allows the book to flow much like an extended oral narration, with long segments of interviews being presented around a series of recurring themes: l’ammedia (envy, or the evil eye), the belief in (consciously or unconsciously) being able to cast a curse through a malevolent glance; the ways in which one becomes a witch or is cured of being a witch; the characteristic behaviors of witches (night flight, shape shifting, dislike of children, compulsion to count); and, of course, the sophisticated system of protection and defense against witches. [17]

Most striking to us, however, is the importance of sound in Volare al sabba, and, in particular, the ways in which oral testimonies are so generously embedded within Bermani’s interpretation, appearing on the page as large, lovingly transcribed sections in dialect, with Italian translations in the footnotes—creating a wall of thick dialectal transcriptions that the average reader will need to toggle away from in order to read the translation. The choice to highlight the linguistic opacity of a defiantly emic perspective is almost (but not quite) akin to the strategic exclusion of non-indigenous readers staged by scholars such as Audra Simpson and Dylan Robinson. [18] This heightened attention to oral testimony is also what most struck us as different from the obvious precedent and inspiration for Bermani’s work—Ernesto De Martino’s seminal research on southern Italian mourning rituals and tarantism in the 1950s, which were of a more traditionally interpretive bent. [19] Bermani aimed to continue this work but also show that, unlike what De Martino had postulated, magical practices were alive well into the 1970s. In other words, De Martino’s implicit attribution of southern magic to a kind of pre-capitalist, agrarian economy was not the whole story and not entirely accurate. We are in broad agreement with Bermani here, but perhaps there is more: the figure of the witch is closely tied to moments of traumatic socio-economic transition—such as the passage from common land to land enclosures, the distinction between salaried and unsalaried labor, and transition from an agrarian to an industrial economy.

Our investigation focuses on two central questions to do with sound and listening that constitute the witch in these transitions. These questions are: in tapes of interviews about witches, what is the relationship between the obviously aural nature of the recording and the role of sound and hearing in the emerging local acoustemology of witches? And what relationship is there, if any, between the methods and technologies of oral history and the experience of witchcraft as narrated and performed by interviewees? It is essential, as a first step towards answering these questions, to say that the discourse of the witch is intrinsically suited to oral history because it is, constitutively, a privileged discourse of hearsay: witches almost never present themselves as such and are almost never seen in action. [20] It is hearsay that makes the witch, that points her out, that tracks her movements, and suggests remedies against her powers. Hearsay indeed deserves a full methodological account in relation to European witchcraft—one that is currently lacking elsewhere, and which we begin to sketch later on in the essay. For now, however, and before we delve into some of the tapes’ contents, suffice it to say that we use the term hearsay in a deliberate, but non-derogatory manner. Hearsay is now a term generally used for discredited, unreliable second-hand evidence, partly because of its historical association with the systematic legal persecution of anomalous behavior, especially witchcraft trials in the Counter Reformation. Yet in the history of witchcraft in Zaccheo and neighboring villages, hearsay requires a more flexible, generative approach that includes not only the signaling of anomalous behavior or even attribution of malevolence to vulnerable people, but also the reabsorption of such behavior into the community.

In other words, hearsay has a powerful role in knitting the witch into (and not out of) the social fabric, while also drawing attention to the social inequalities that foster an environment conducive to the figure of the witch. This communitarian function of hearsay brings up important questions regarding the effective reach of Counter Reformation heresy trials. It seems significant that Zaccheo and the neighboring area was, to our and Bermani’s knowledge, never directly a site of persecution of witches during the Counter Reformation—Zaccheo is, to this day, a relatively isolated part of the region, one that Counter Reformation officials didn’t directly control. [21] As such, witchcraft was perhaps not as closely associated with a history of trials or even persecution, but rather it was a means of internally managing political tensions in a largely non-violent manner. We will argue, indeed, that hearsay, in this world, has a particular power to summon the unfathomable to the senses without calling for its destruction.

Why the Word “Witch”?

No discussion of witchcraft today can begin without the acknowledgement that the very term “witch” is controversial and an evolving matter of discussion in anthropology and history. [22] Terms such as “magic” and “witchcraft” were crucial to the formation of the field of anthropology in the early twentieth century, where comparative approaches—often based on second-hand fieldwork—were essential to establishing a set of shared methodologies for cultural practices that eschewed the bounds of Western medical and legal practice. [23] Such broad comparativism came rightly under scrutiny with the 1970s turn towards subaltern and area studies, which focused on local practices, emic language, and a suspicion of universally applied terminology as an a-critical, Eurocentric approach. Within this latter turn, which is still dominant not only in anthropology but in the humanities writ large, terms such as “magic” and “witchcraft” lose descriptive power precisely because, when used in broad intercultural sense, they risk describing anything that escapes an implicit Eurocentric notion of reason and science. From an historical standpoint, particularly as it concerns the history of European witch hunts in the Counter Reformation, it has been argued that the very term “witchcraft” reproduces the Counter Reformation’s flat, misogynistic, and persecutory gaze on the multitude of local community practices and medicines that the term encompassed at the time of the trials. [24]

In this paper, we are using the terms “witch” and “witchcraft” deliberately, and for a number of interconnected methodological reasons. For one, “witch” is the translation of the dialectal term used by Zaccheo’s inhabitants, sdraje—which is the Teramo dialectal form of the Italian “strega/streghe” ( sdraje functions as both singular and plural noun). The Italian noun is tied to the Latin “strix,” for nightbird, and is generally translated as “witch” in English. We could have insisted on using the term sdraje throughout, but resisted this because we believe that the term requires a mindfully comparative approach that takes stock of the braided continuities of the figure of the witch in a European context and perhaps beyond. Indeed, despite a widespread postcolonial turn in the humanities towards historical and geographical specificity and emic (rather than etic) language, witchcraft is a topic that has invited a comparative approach even from historians and anthropologists. Such comparativism is done nervously and with a richness of disclaimers, but also with a justified sense of methodological urgency. [25] This is in large part because witch hunts in Europe coincided, and overlapped, with colonial conquest and the denigration of indigenous religious practices as witchcraft. Silvia Federici’s seminal argument in her Caliban and the Witch: Women, The Body And Primitive Accumulation (to which we will later return) is that the category of the witch marks a moment in the sixteenth century in which the subjugation of women and the mechanisms of settler colonialism become part of a unified effort to secure land as private property. [26] Methodologically, this means that witchcraft is a significant and even unique point of contact between the disciplines of history and anthropology and a generative topic for the new strains of careful comparativism embodied by interdisciplines such as global history.

These considerations inform one of the most recent, broad-ranging books on this topic, British historian Ronald Hutton’s 2017 The Witch: A History of Fear, from Ancient Times to the Present. [27] Hutton’s approach is one of open, inquisitive, carefully documented comparativism, in which the category of “witch” clasps together a series of political and historical figures that are, as he shows, recurrent across a wide range of societies and whose emergence is neither predictable nor easily reducible to a set of social and historical factors. After taking into consideration sources spanning from Middle Eastern and Mediterranean traditions, as well as drawing from a wealth of anthropological work on cross-cultural understandings of witchcraft, Hutton highlights four core features that ground a comparative analysis of the figure of the witch:

The first … was that such a person worked to harm neighbours or kin rather than strangers, and so was an internal threat to a community. The second was that the appearance of a witch was not an isolated and unique event. Witches were expected to work within a tradition, and to use techniques and resources handed down within that tradition, acquiring them by inheritance, initiation or the spontaneous manifestation of the particular powers to which they were connected. The third component of the European stereotype of the witch was that such a person was accorded general social hostility, of a very strong kind … Finally, it was generally agreed that witches could and should be resisted, most commonly by forcing or persuading them to lift their curses; or by making a direct physical attack on them to kill or wound them; or by prosecuting them at law, with a view to breaking their power by a punishment which could extend to having them legally put to death. [28]

It is remarkable how, even from the short opening episode of Francesco’s story, we can get a sense of the applicability of these four criteria. First, we see the importance of witches as a community-based figure—one that lives in a specific locality and harms only those within it (the crossroads for meeting witches are all within the bounds of the village). Second, the importance of tradition—as highlighted by the women’s discussion of the emergence of new witches in Zaccheo. Indeed, as Bermani shows, there were many codified conditions under which one might be born a witch or become one. [29] Third, the witch was feared—so much so that confronting one, as Francesco did, was considered the ultimate show of masculine strength. Lastly, the need to resist the witch by either persuading her to lift the curse, performing protective counterspells, or by quickly scolding her with a hot poker in order to turn her back into a friendly figure. It would be disingenuous to say that the latter act wasn’t a form of physical disciplining. Yet it is also important to note that, in the community, this was considered a very quick and not especially violent act, often performed on children as a form of immunization against a range of evil spirits. [30] Such practices signal that coexisting with witches was an accepted part of everyday life.

Lastly, a broader comparative approach is essential to understanding the political dimension of the acoustemology of witches in the Zaccheo tapes. It is striking that a feminist Marxist historian such as Federici considered Calibana preparatory study for the examination of a return to witch hunts and femicide in the global twenty-first century. [31] As mentioned earlier, Federici argues in Caliban that the process of land expropriation and privatization in both North and Latin America and Europe was accompanied by the emergence of a salaried, male, and white labor force to the devaluation, persecution and economic exclusion of all other forms of labor. Unpaid labor, such as feminine reproductive labor in Europe and plantation slavery in settler colonies were part of the same system for drawing profit from cheaply-worked, stolen land, and both forms of unpaid labor were tied to accusations of witchcraft. Federici’s take on the phenomenon of witchcraft is strikingly broad to this day, and it is an implicit response to a focus on the hyper-local (the village) when discussing this topic. [32] We will begin to consider the applicability of this history of land privatization to Zaccheo below, but for now may it suffice to say that we share Federici’s ambition to discuss witchcraft as a response to epochal socio-political shifts. The creation of the witch in moments of historical transition corresponds to a sensorium closely attuned to failing sight, dangerous listening, and sounds without a detectable source.

Performative Denial and Nurtured Hearsay

In considering the discourse of the witch in Zaccheo, we highlight three key dimensions of hearsay. The first is that the suspected witch denies, when questioned, all knowledge of witchcraft, but then feeds hearsay by doing things she knows will identify her as a witch. The second is a constitutive inability to see witchcraft being performed, and the remedial use of hearing. The third is connected to the witch’s own hearing abilities: on Fridays (and other days, according to different traditions), she is able to hear when people talk about her, creating a reversal of the atmosphere of surveillance and suspicion that makes the witch in the first place.

Let’s now consider the first dimension of hearsay: denying allegations of witchcraft while performing behaviors usually attributed to the witch. In the years of Bermani’s research there was one witch uniformly pointed out in Zaccheo: Rina, a low-income, elderly unmarried woman with no children who lived with her sister (also unmarried and with no children). Rina and her sister lived in the village square, around which most of the houses were located, adjacent to one other. Bermani recorded Rina only once—in her home in 1966—together with her sister and a younger woman present, Gloria, who translated between dialect and Italian. When pressed by Bermani on whether she knows about witches or about people’s gossip about her, Rina responds by saying she doesn’t know anything about such beliefs:

Rina declares that she knows nothing of witches and she doesn’t go around sendenne (hearing/feeling/hearing about) things; the act of refusal is similar to that of the women surrounding Francesco earlier, all of whom declare that they do not believe in witches. As with those women, Rina’s denial is not an ending; instead, it is a necessary premise to a different mode of conjuring the witch. In the case of the women encountered at the opening of the essay, this conjuring was communal dialectal speech; in the case of Rina, it is the performance of a series of small behaviors that she knows will generate gossip about her. Many of Bermani’s interlocutors, as well as many who are still living in Zaccheo to this day and with whom we have spoken informally, remember how Rina would perform small actions that, while harmless in themselves, were interpreted as symptomatic of being a witch in Zaccheo. [34] For example, she was unwilling to kiss children when asked, and she wouldn’t leave a neighbor’s house when an upside down broom had been placed by the exit. These actions carried significant weight: having a suspected witch kiss one’s children constituted an act of protection against her curse, so this is something that a witch would avoid doing unless she was either threatened or bribed. The upside down broom trick was a well-known way of trapping a witch in one’s house until she agreed, for instance, to lift a curse. [35] Therefore, while Rina openly denies any involvement with sendè of witches, she is in fact openly feeding it in her everyday life. How, and why, would she do such a thing?

The answer lies in a mechanism for the generation of hearsay, and the role of hearsay in securing—by generating fear for the witch—the physical and social survival of elderly, unmarried women. Behaving, as Rina does, in ways that will generate gossip is the second part of a ritual that begins with denying all knowledge of witches. On the same day of her interview with Bermani, we learn about Rina’s undeniably witchy behaviors when he converses with another group of women in another home. This gathering includes Maria, who’s hosting the gathering at her house, as well as Gloria (the young woman present on the taped interview of Rina), who leads the conversation:

The recording is amazing because of the delight, laughter, and warmth that emerges in the joint pleasure of discussing Rina’s behavior. Laughter occurs fairly often in Bermani’s interviews, and it is common for it to be laughter among a group of women. We heard it already around the kitchen table where Francesco was narrating; and its valence, in our opinion, is similar across these two recordings. In other words, this is not a malevolent laughter aimed at an outsider, but an expression of glee at discussing something forbidden. By laughing, the speakers mask themselves in a moment of reticence—they withhold their speech from both microphone and interviewer—while expressing their participation in a discourse that gives them pleasure. Laughter tells us that witchcraft is a sensitive topic (hence the need to giggle and screen oneself off from speech) that has a function of catharsis, a release of pent-up psychic energy. [37] In the story told by the women, Rina commits two defiant acts: openly admitting to not liking children and urinating on the floor. This latter act was connected to the belief that, if a witch is blocked in someone’s home after midnight, her only way out is to pee on the house floor. At the time, indoor bathrooms were almost nonexistent in Zaccheo, and people would need to relieve themselves outside. By peeing indoors, the witch inverts the inside/outside the house social norm, thus freeing herself. It is to this ritual of inversion of social norms that the women are reacting, and they react with glee, vicariously participating in, rather than rejecting, a behavior that attracts them and repulses them at once.

Rina is therefore an active participant in, and not a passive victim of, the mechanism of hearsay. As an unwed, child-free woman, Rina has a social function to absolve: she consciously becomes a conduit for antisocial, unconventional behavior—of a kind that other women must repress in public. In return, she (not having a husband or children to support her and, being a woman, barred from salaried work) can ask for food and practical help from her neighbors without being denied. Amidst the brutal inequalities brought on by a rural society centered around property and salaried male labor, her role as a witch doesn’t signify, in the community, a radical correction of that inequality. Instead, it’s an acknowledgement of a vestigial sense of communal obligation that comes around, literally, knocking on people’s doors. Both in Bermani’s interviews as well as from our own research, [38] we have learned that Rina’s means of survival derived from traveling on foot around Zaccheo and the neighboring villages, going door to door asking for daily work as well as charity in the form of flour, oil, and other staple foods. The mechanism of hearsay is tied to the political economy and social inequalities of Zaccheo and its environs. Rina must deny all knowledge of witchcraft and yet behave publicly in ways that will make people talk. This talk, in turn, lends Rina a form of social capital that allows her to be woven into the village’s social fabric and thus survive.

Hearing the Witch: The Work of Acousmesis

Now for the second dimension of hearsay, which we defined above as the inability to see a witch in action and, instead, of hearing/feeling her. Even Francesco—who saw several women at the crossroads whom he recognized as witches—could not catch them performing witching acts such as flying, braiding horses’ hair, and casting curses. In fact, hearing/feeling the witch is a recurrent trope in the tapes: several people say they can hear and feel her, but nobody can see her in action. This, as we have seen, is partly because she operates at night, usually in poorly lit parts of the town. Yet the witch’s invisibility is more deeply built into her ontology than mere circumstance. In sound studies parlance, we could say that the experience of the witch is acousmatic—that is, that she takes the form of sounds whose source is occluded from the senses but also semiotically unplaceable. Michel Chion famously articulated the particular power held by cinematic voices whose sources are invisible; and it is perhaps no coincidence that, in doing so, he conjures a witch-like creature: “everything,” he notes, “hangs on whether or not the acousmêtre has been seen. In the case where it remains not-yet-seen, even an insignificant acousmatic voice becomes invested with magical powers as soon as it is involved, however slightly, in the image. The powers are usually malevolent, occasionally tutelary.” [39]

We can explore the notion of the acousmatic in more depth by considering an interview from August 16, 1966. Gerardo, a 34-year old resident of Zaccheo, recalls encountering a witch in the form of crying sounds outdoors. He tells his story without hesitation, while his father is listening and confirming, at the end, that what his son witnessed was the presence of a witch:

Three aspects are crucial here. For one, the sound is deemed strange not just because it has no indexical referent (it sounded like a child weeping) but because it is occurring at an unusual time and place (nighttime and away from houses). The sound is also intermittent and difficult to position in space, and therefore invites heightened attention. Secondly, Gerardo’s trick of throwing a stone towards the source of the sound marks the gap between the sound and its source, measures it, and therefore makes that gap real, materially expressive. Lastly, the flare of fire and temporary blindness as the witch escapes doesn’t resolve, but rather draws new sense from the inability to see the source. In other words, Gerardo’s encounter hinges not so much on poor visibility (lack of light, nighttime) but on the experience of temporarily losing one’s ability to see. The relationship of acousmatic listening to witchcraft goes from being circumstantial (it was nighttime, I couldn’t see) to being constitutive and necessary (she called to me and then blinded me as she escaped, so I could actively see her in the act of vanishing). Here the acousmatic is neither a mark of an omnipotent other, nor a semiotic disturbance awaiting resolution—but instead a call, an invitation to hearken to, to measure (the stone’s throw), and finally to behold (the blinding flare) the disappearance of the sound’s source. Equally striking is that the disappearance-through-temporary blindness coincides with the reconnection of the witch’s voice with her body.

We can observe a similar experience in the testimony of Omero, interviewed by Bermani on August 29, 1969. Omero remembers going, four or five years earlier, to a particular crossroads in a neighboring village where witches transited at night. This was in order to help a neighbor’s children who were often sick and thought to be tormented by witches (like Francesco’s daughter Lillina, previously discussed). Omero agreed to help by meeting the witches at the crossroads at midnight, as he had already done for his own daughter. Night-time, let us remember, meant no visibility, especially in the country lanes that wound around the crossroads where Omero went. He tells that the first time he went, he also brought a scythe for protection; but in so doing, he broke the protocol for witch visitations, and scared the witches so much that they changed course and didn’t pass by the crossroads; Omero remembers hearing them clamor nearby:

The second time, Omero left the scythe at home, and so could watch the witches transit. And yet, he said, once again couldn’t truly see them: “I couldn’t recognize them, because they were all veiled. I could see the horses, but I couldn’t tell if they were women or men.” [42] While he was able to distinguish the horses (both tall and short, all thin) and noticed colorful clothes, he couldn’t give any details about the witches’ appearance gleaned from sight. Rather, the witches took (sometimes unexpected) shape through sound; and it is Omero’s inability to see that provides the condition of this sonorous embodiment, as his attention focuses on the terrible noises that announced the witches’ arrival.

Other people interviewed by Bermani similarly experience the witch not so much as invisible, but as a momentary lapse in vision, and such temporary blindness corresponds to the power, specific to witches, to shapeshift.Let’s now turn to an extraordinary recording from 1966—a witness account of shapeshifting, provided by Gerardo, whom we have encountered above. The protagonist of the story is, once again, Rina, this time accompanied by her sister Giovanna:

It’s hard to put into words the vivid soundscape of this tape. Listening to it, we bear witness to Gerardo’s talent for storytelling—setting an evocative scene, the use of suspense, the presence of sound effects—as well as the pleasure of sharing such an extraordinary occurrence. Importantly, we can see how, in telling stories of hearing witches, the acousmatic realm is created through coordination of narrative and sound effects provided by the teller (“VOOOM!”). In fact, the acousmatic becomes almost a Barthesian “reality effect” in stories about witches: the stretching of the gap between sound and source is the place where the witch manifests herself, in the most material sense possible, in a momentary lapse of vision and understanding. [44]

The above examples and analyses suggest that Chion’s classical definition of the acousmêtre requires some revision for the purposes of an acoustemology of witches. For Chion, the acousmêtre is a speaking, intelligible voice that addresses people in the cinematic frame. The witch in Zaccheo is never addressed directly and doesn’t ever speak to others in her role as a witch. Indeed, as we saw from Francesco’s story, addressing a witch at a crossroads at night results in death. The acousmatic sounds that signal a witch’s presence are not her voice or anything directly traceable to her physical form, but genuine malfunctions of the sensorium, strange sounds without an obvious source that signal a dangerous (and witch-related) lapse in audio-visual perception. Brian Kane, in his Sound Unseen , argues, against the French intellectual tradition of acousmatic listening, that the gap created between a sound and its source is not in itself that significant an event. [45] A sound isn’t made powerful by its source being occluded; instead, it is the cultural techniques and discourses that fill that gap that make a sound acousmatic, that lend it power in a sociological and political sense. In other words, for Kane, it is listeners who create—through discourse, and for specific socio-political reasons—an acousmêtre. In the case of witches in Zaccheo, sounds that come from no-one, or have no reasonable source, are a lightning rod for a listener-generated discourse of hearing/feeling the witch without seeing her. This discourse, as we will see below, has political and historical origins in the impossibility of ocular witnessing in witch trials; and it has developed, in Zaccheo, into a form of creative hearsay whose impact on the shared sensorium is profound.

The Witches Can Hear, and Speaking “As If”

The third dimension of hearsay that we want to highlight is the ability of the witch to overhear what is said about them on certain days, depending on each family’s inherited belief. Being able to hear people talking about her, for the witch, was a way to identify and punish those who might give her away, and so endanger her powers. Tuesdays, Wednesdays, and Fridays were all days on which witches’ ears for gossip about them was especially keen. In Zaccheo, people might deem either Tuesdays or Wednesdays dangerous for discussing witches, but Friday was a day on which witch-talk was uniformly understood to be forbidden. In several recordings, we can hear people double-check the day of the week before launching into narrations about witches; other times the ascertaining of the date is not uttered, and people simply speak and act in accordance with the day of the week. In these instances, interviewees need to gauge what they want to be heard or not heard saying on the microphone; and when they don’t want to be heard, they start speaking softly, turn their faces away from Bermani and talk amongst themselves, speak in dialect only (which they know Bermani cannot always understand). [46] Such a protocol of reticence indicates that the tape might have been recorded on a day on which it was forbidden to speak of witches. The opening episode of this essay—Francesco’s story about visiting the crossroads and the ensuing conversation among his daughters and Mariafelice Forti—was recorded, according to Bermani, on a Wednesday. The interviewees’ awareness of witches eavesdropping on the conversation might well be the reason why Francesco does not identify the witches by name. It might also explain the opaque quality of the women’s ensuing discussion regarding the survival of witches in the area. What is more, in examining the tapes, we noticed a particular acoustic quality to some of these recordings. Namely, there are times in which Bermani raises the gain of the recorder to the highest level in order to capture these reticent voices, making the recording full of background noise and prone to sudden distortions. This, as much as the interviewees’ behavior, signals the particular dynamic of the interviewer and interviewee on a tape recorded on a forbidden day.

We’re going to turn our attention to one such tape from August 1964. Here, participants are reticent and the recording is confused, while, on other tapes, the same people had no qualms talking to Bermani about witchcraft. Although we don’t have an exact date, this interaction has the feel of a recording made on a forbidden day. Here, Bermani is asking about witches and the several women present answer more and more evasively until someone reminds the others: “he is already recording.” After some consultation, the women decide to sing a song in three-part harmony, given that they knew Bermani’s main focus at the time had been popular songs:

In this recording, the singing at the end is so loud, given the high gain of the recording, that it sounds distorted. There is an interplay between the presence of a sonic record and the ways in which people tactically withdraw from the microphone without, however, being silent; and this interplay is an important part of the acoustemology of witches in oral history. On the one hand, the interviewees seem to be resisting the extractive practice of recording bits of conversation. The switch to dialect is one such act of micro-resistance against their interviewer. It is also remarkable that the interviewees don’t flat out refuse to discuss witches, but do so in a hushed voice. Their refusal is performative, it is meant to be witnessed and heard by others. This, then, leads us back to the question of hearsay: on a Friday (or Tuesday or Wednesday), one must be heard speaking of witches as if one didn’t wish to be heard. Interviewees perform their withdrawal from the microphone and from the witch’s ear by marking their own speech as forbidden. They lend power to their speech by using it sparingly, suspiciously—but continuing to use it all the same. In this sense, the technology of the microphone plays into, not against, the acoustemology of witchcraft. On days in which witch-talk is forbidden, people must speak as if under surveillance, and be heard to do so. The microphone becomes an extension of the ubiquitous hearing of witches, and as such it doesn’t foreclose witch-talk, but lends it special power precisely by monitoring it, by recording the reticence with which it is offered. As an extension of the witches’ keen ear on certain days, the microphone participates in a ritual poetic justice: it allows townsfolk to project back on themselves the monitoring and gossiping—the generation of hearsay evidence—that served to mold the figure of the witch. [48]

Hearsay and the Work of “Darknesse”

Does witchcraft possess a codifiable relationship to sound and hearing? And if so, how might we trace such a relationship and how does it manifest in the case of the recordings in Zaccheo? In 2004, anthropologists Pamela J. Stewart and Andrew Strathern devoted a substantial volume—including scholarly perspectives on African countries, India, New Guinea, North America and Europe—to the particular, and to their minds, overlooked relationship between the study of witchcraft and the study of rumor and gossip. They justified their commingling of these two anthropological fields of inquiry as follows:

Ideas about witchcraft and sorcery have often been pointed to by anthropologists and social historians as markers of social stress (notably by Marwick 1965). Our point is that rumors and gossip enter into the early stages of the development of stressful circumstances and so lead into later stages that may crystallize in accusations […]. Witch trials represent the confluence of … local and interlocal events, culminating in acts of expurgation or scapegoating, usually directed against those who are socially weak or marginal. [49]

For Stewart and Strathern, the relationship between gossip, rumors, and witchcraft passes by way of witch hunts and trials, which are often the culmination of a process of collective suspicion towards, and then formal accusation against, a set of vulnerable individuals. It is important to clarify that, in the context of Zaccheo, the “culmination in expurgation and scapegoating” through witch hunts and trials did not occur. Nevertheless, here too, the element of rumor and hearsay was foundational to, and generative of, the figure of the witch. This prompts us to consider the function and result of rumor and gossip aside from their “crystallization into accusations.” Indeed, once we lift the cause-and-effect connection between rumors and witch hunts, the point about the relationship between hearsay and witchcraft can be refined into an acoustemological inquiry, particularly within a European context.

The acoustemological roots of the witch might be found, within European history, in the slippery, emergent legal distinction between direct (ocular) witnessing and indirect witnessing, and the special use of indirect witnessing where a direct (ocular) witness isn’t available. Early modern witch hunts caused a breakdown and then rebuilding of legal systems built around witness statements and the evaluation of these statements’ validity—a process that left traces far beyond the seventeenth century and the realm of witchcraft. It is indeed because of the horrific witch hunts in the European and American sixteenth and seventeenth centuries that we have an explicit record of the evolving and complex role of rumors—community-based, indirect witnessing, hearsay—in court law, with far-reaching implications in legal history and much more broadly. [50]

The distinction between direct ocular witnesses and indirect witnesses in Roman Law corresponded to the de visu/de auditu distinction, according to which direct and indirect witnessing are implicitly (and messily) codified through the Western hierarchy of sight (understood as stable, objective, and rational) over hearing (understood as unstable, subjective, and emotional). We might say that the distinction between acceptable direct (ocular) witnessing and everything else—whether it is a direct aural testimony with no visual confirmation, or a fully indirect witnessing—presents a sliding scale between the ocular and the aural. In this paradigm, at the strongest end of the credibility spectrum, we have direct witnessing, which has to be ocular; this is not because hearing doesn’t come into it (one may well see and hear someone do something) but because it is the eye that determines the true source of a sound. When unable to see directly, a witness might recognize someone’s voice saying something; this, however, isn’t the same as seeing the person talk, and therefore direct aural testimony without visual confirmation is far less valuable as evidence. [51] This hierarchy of eye over ear is what Jonathan Sterne famously referred to as the “audio-visual litany”—the critical cultural history of which is, in many ways, the bedrock of sound studies and acoustemology as a discipline. [52]

Hearsay, like hearing, certainly carries a negative bias against the aural realm as a site of rational knowledge in Western thought (a bias that, of course, corresponds to what Jacques Derrida diagnosed as the overvaluation of voice as origin and metaphysical truth over writing). [53] Yet hearsay is also tied to a particular history of the state of exception that allowed rumor to stand as valid testimony, in situations in which ocular witnessing was deemed impossible and indirect witnessing (with its aural implication) became the next-best alternative. In fact, the legal history of witchcraft shows that the negative bias against hearing and voice could become twisted into a positive bias in a violent state of exception.[54] In other words, witchcraft and the law of hearsay are not just related, but messily and chronically co-emergent states of exception, in which gossip and indirect witnessing came to have the legal weight of an accusation that could be sustained in court. The point is crucial lest we fall prey to a simplistic view of early modern witch hunts as being symptomatic of a pre-modern, and implicitly unreformed and pre-rational, understanding of the law. On the contrary, the rejection of hearsay as proof in Roman Law as well as English Case Law was categorical, and so closer to contemporary legal attitudes towards hearsay (according to which hearsay is admitted in court only under a highly restricted set of circumstances). [55] The use of hearsay in court is, therefore, not so much a thing of a pre-modern past, but a chronic political and historical condition activated in moments of crisis.

We see this outlined with hair-raising clarity in a famous manual for sixteenth-century witch hunts, the Malleus Maleficarum. [56] Here, the three means of initiating an inquest into witchcraft are outlined like so:

The first question, then, is what is the suitable method of instituting a process on behalf of the faith against witches. In answer to this it must be said that there are three methods allowed by Canon Law.
The first is when someone accuses a person before a judge of the crime of heresy, or of protecting heretics, offering to prove it, and to submit himself to the penalty of talion if he fails to prove it.
The second method is when someone denounces a person, but does not offer to prove it and is not willing to embroil himself in the matter; but says that he lays information out of zeal for the faith …
The third method involves an inquisition, that is, when there is no accuser or informer, but a general report that there are witches in some town or place; and then the Judge must proceed, not at the instance of any party, but simply by the virtue of his office. [57]

Note how three modes of accusation are listed in reverse order of verifiability and accountability, but also, as becomes apparent, in increasing order of legal and theological value. Indeed, the first option (involving someone making themselves accountable for false witnessing) is “not actuated by motives of faith;” importantly, it is not “very applicable to the case of witches, since they commit their deeds in secret.” [58] By the time the third method is appraised, we have gone past indirect witnessing into a regime of paranoid listening in which the inquisitor may act in service of the true faith:

The third method of beginning a process is the commonest and most usual one, because it is secret, and no accuser or informer has to appear. But when there is a general report of witchcraft in some town or parish, because of this report the Judge may proceed without a general citation or admonition as above, since the noise of that report comes often to his ears; and then again he can begin a process in the presence of the persons, as we have said before. [59]

This passage goes beyond the notion of hearsay as a necessary if unpleasant substitution to direct ocular witnessing and into hearsay as the preferred method of accusing a witch whose work cannot be seen. Indeed, the ocular is almost discounted, or certainly not declared preferable anymore.

In a strange twist of the Pythagorean definition of the acousmatic, here it is the unseeing eye that unblocks the ear as a site of knowledge. In other words, hearsay is accompanied by an intense feeling of unseeing. It is the active obstruction of sense—of seeing that one does not see—that works as the trigger for the ear as a compensatory site of knowledge and guides the processing of aural stimuli. This mechanism was at work in Western legal systems more broadly, which betray an almost existential obfuscation of sight in the face of witchcraft. In 1618, one of many legal pamphlets published in England on best practices for witch hunting warned “not alwaies to expect direct evidence [against witches], seeing all their works are the works of darkenesse, and no witnesses present with them to accuse them.” [60] The existential and political relationship of witchcraft to hearing and hearsay must go through a political phenomenology of a “darknesse” that makes anxious hearkening necessary.

Towards an Acoustemology of Sendè

We return, finally, to the Zaccheo tapes. What, exactly, is the relationship we are positing between emergent legal protocols in Early Modern Europe and hearsay in a town in twentieth-century Abruzzo? Certainly, the practices attributed to witches in Zaccheo correspond to the behaviors listed as witchcraft in sixteenth-century demonological works: the harassing and kidnapping of children and livestock, petty acts of cruelty motivated by envy, night-flight, the ability to transform into an animal, anti-social behavior (living alone, disliking children, asking neighbors for food donations). Abruzzo as a region was a site of fierce inquisitorial disciplining, with documented witch hunts and burnings between the sixteenth and early eighteenth century in Penne, Chieti, Teramo, and several other places. [61] As we have seen, however, Zaccheo and its neighboring villages were not sites of persecution. It is plausible, then, that Zaccheo picked up some of the sensorium associated with witch hunts but did not associate it with reporting to a disciplining authority; instead these practices were folded into a fragile, internally resolved community equilibrium.

This is not to say, of course, that Zaccheo offers a happy anomaly or an historically innocent version of witchcraft. Rather, it shows the complex role of “darknesse” and the urgent recourse to the ear in communities that were affected by it. In their aforementioned work, Stewart and Strathern note that gossip and rumor are responses to acute societal stress; Zaccheo, at the time of Bermani’s recording, was a place where land was owned by wealthy landowners (Mariafelice’s father, Bermani’s father-in-law at that time, was one of them) and worked by salaried farm hands (Francesco, who supported his wife and six children through his labor in the fields, was one such person). Between male salaried labor and landowners were petty landowners, that is to say families who managed to buy themselves a piece of land to work on their own, or families who owned agricultural machinery and worked other people’s land. Women, though crucial to the community, did not produce value in this economy because they neither owned land nor worked the land for a salary, and were mostly involved with managing the home and children. Although the state provided a small pension for people without a means of subsistence, this money was often insufficient to cover basic expenses such as rent (unwed women were rarely homeowners). This vulnerable status of women in the transition between agrarian and capitalist economy is, for Federici, the breeding ground of witchcraft:

Particularly disadvantaged were older women who, no longer supported by their children, fell onto the poor rolls or survived by borrowing, petty theft, and delayed payments. The outcome was a peasantry polarized not only by the deepening economic inequalities, but by a web of hatred and resentments that is well-documented in the records of the witch-hunt, which show that quarrels relating to requests for help, the trespassing of animals, or unpaid rents were in the background of many accusations. [62]

Petty theft, particularly of farm animals, land trespassing (witches often manifested in the roads between properties) borne of resentment were indeed an important part of the profile of the witch in Zaccheo, even if, as mentioned several times already, these incidents mostly did not result in persecution and prosecution. Marginalized, unwed, child-free and non-earning women, such as Rina, were likely to be blamed for such misdemeanors and so cast as witches. Yet, in some cases, this casting was not simply a matter of scapegoating, but also the thing that ensured the woman’s livelihood. The fear of a witch’s reprisal when denied a favor expressed, as Federici would argue, a barely repressed awareness of the brutality of private property and salaried labor. This is precisely the role that witchcraft played in Zaccheo: it articulated a fear of those not accounted for by family and property, while simultaneously protecting those same, unaccounted-for people from penury.

It is no coincidence that so many of Bermani’s recordings were made around kitchen tables—communal, feminine spaces where women might gather to work together on home-related tasks: food preparation and preservation, mending, and so on. These spaces were full of talking, discursively alive: multiple children and multiple generations of women were almost always, and often disruptively, present on tape. [63] In this context, witches were discussed through the slipstream of sendè— of hearing, hearing about, and feeling, often in contradiction of an outward dismissal of witches. Sendè, in other words, signaled a perilous but necessary reliance on hearing and hearsay in a world of existential darkness: shifting property relations, the loosening of female kinship, an altered relation to one’s rights, one’s belongings, and coping with new forms of social inequality. In its more positive connotation, sendè is a creative network of hearsay that binds—through a series of sometimes friendly, sometimes threatening rituals—the witch to her community, and they to her.

There are, evidently, many methodological gaps to be filled in this initial set of considerations on doing an acoustemology of witchcraft through oral history, in Zaccheo and perhaps elsewhere. For one thing, it remains to be studied how land privatization and salaried labor (with the attendant social inequalities between the earning and unearning) emerged in this area and by what means it was sustained into the 1960s, the decade in which Bermani’s interviews took place. It might be that, in rural areas such as Zaccheo and environs, the middle stage between agrarian economy and industrial capitalism (as described by Federici) became an extended transition, essentially, over more than four centuries; and that, over this time, the figure of the witch evolved from a figure of persecution into something that needed to be folded into a longer-term equilibrium. It is telling, in this respect, that the crossroads where one could go and see witches were often at the boundary between lands owned by different people, marking a place quite literally in between properties at a time when property had come to determine societal value. It is by these same roads that land workers traveled to and from their jobs working private land; and by these same roads that Rina, a lone woman expected to stay home, traveled on foot from village to village to ask for alms. Bermani himself visited the village towards the end of this long period of transition, as finally the landowning classes became less tied to their land and the surrounding area was given over to industrialization. Indeed, many of the crossroads indicated by people in their interviews with Bermani are now gone, taken over by factories that drew more people to the area, and which are now, in turn, in crisis due to deindustrialization and outsourcing of labor elsewhere.

Listening to the Zaccheo tapes, to the voices of those experiencing seismic socio-economic changes taking place in the Italian 1960s, we bear witness to the ways in which being unable to see—and the recourse to hearing and hearsay that arose in their wake—were not only responses to stress or the manifestation of superstitious beliefs. The witch was not only a local, unfortunate, and persecuted byproduct of world-historical forces that conspired to enclose common lands. She was also, in an anthropological sense, its incarnation and material expression. Her complex embodiment bodied forth the ways in which a world lost concreteness, could be reversed and concentrated into a tangible figure. The witch was made in the anxious hearing that confronted darkness, during a time in which a cry in the tree or clamor at a crossroads could replace what one failed to see.