Interview History Text Interviews

Invisible constraints
Interview with Carlo Ginzburg

by Florent Brayard , 15 December 2010
with the support of FMSH

Télécharger l'article : PDF

Carlo Ginzburg, a historian specialising in witchcraft and popular beliefs, has produced an exceptional range of rich and powerful work. In this interview he looks back at his intellectual development, which has taken him from the Friulian countryside to the conspiracies of the 1970s and 1980s, and discusses the personal aspect of his relationship with the archives.

Carlo Ginzburg was born in 1939 yet comes across as a young man. When he talks, his speech is punctuated by exclamations like “I couldn’t believe it”, “I was stunned”, “shocked”, “fascinated”; he uses words like “pressure”, “struggle”, “challenge”, “evidence”. He is one of the most influential historians of his generation and probably the most committed to his work. Ginzburg has worked on a wide variety of subjects yet there are clear themes running through his work. He has held teaching positions at the University of Bologna and at the University of California Los Angeles (UCLA); currently he teaches at the Scuola Normale Superiore in Pisa. Ginzburg made a name for himself through his research on witchcraft and popular culture in the sixteenth century which he has published in a series of outstanding books: The Night Battles, The Cheese and the Worms, and Ecstasies: Deciphering the Witches’ Sabbath.

For twenty years now, Ginzburg’s output has consisted mainly of short articles, or essays. Ranging from the readers of Ovid’s Metamorphoses to Pope John Paul II’s slips, Ginzburg’s fields of interest cover a significant part of the cultural and social history of Europe. Following the success of Clues, Myths, and the Historical Method and Wooden Eyes: Nine reflections on distance, Le Fil et les Traces (Connections and Clues), the French translation of a new collection by Ginzburg was published last autumn. Once again Ginzburg is dazzlingly erudite, the book is a joy to read, and he provides some remarkable insights that challenge the reader’s established views.

The idea of the interview was to discuss the relationship between what we do and who we are: there is a constant two-way flow between an author and his work which is useful to understand so we can better appreciate the final result – what is finally published. Some of my questions were probably unwise, but in any case, here is the result.

La Vie des Idées: In a 1980 interview for the French newspaper Le Monde, you said that naturally, historians don’t write to understand the origins of their life or the context of this life, but that it would be very interesting to psychoanalyse the process of historical research, with examples. Thirty years on, would you still stand by what you said?

Carlo Ginzburg: It would certainly be interesting, but there would be a number of technical problems. Who would speak, the psychoanalyst or the patient? And if it was the patient, how would you know that what he was saying was reliable? You would have a new source, but you couldn’t guarantee its veracity. Basically, you would be back to the old dilemma, that no single document is reliable in itself.

La Vie des Idées: Psychoanalysing the process of historical research could simply be a way of re-reading an intellectual journey, and discovering links between events that had not been made at the time. Let’s take Ecstasies: Deciphering the Witches’ Sabbath (published in 1989) as an example. The book seeks to demonstrate that some elements of the sabbath date back to age-old religious beliefs which have been passed on orally over millennia. In order to do this, you compare quite different phenomena on the basis of formal similarities. The first section of the book however takes a rather different approach from this innovative method – it’s a traditional historical account which…

Carlo Ginzburg: I’m sorry, I’m going to have to interrupt you there. The first section of the book certainly appears more traditional. After that the book moves on in other directions which are related more to morphology (although later on I return to history again). One critic’s review stopped at the end of this first section, and he said how incompetent it was without mentioning the rest of the book. This reminded me somewhat of that fascinating discussion between Robert Fawtier and Marc Bloch in the review Le Moyen Âge (The Middle Ages). Fawtier, a confirmed medievalist, had slated The Royal Touch, questioning whether a large part of the book could really be classified as history at all.

The idea of starting up a dialogue between history and other disciplines has always fascinated me. We often tend to think of history as a fortress: it’s not a fortress, it’s an airport. The desire to engage in dialogue underpins this book, but history is also present too. My aim was to try and determine the relationship between an event and the underlying structure, and so I had to describe how the event happened before investigating the origins of the event far back in the past. There again, in some ways, my approach could be compared to Marc Bloch’s. Bloch was one of the first people to attempt to link these two dimensions in his articles on agrarian individualism and in French Rural History: An Essay on its Basic Characteristics, which I translated into Italian.

The relevance of conspiracies today

La Vie des Idées: In the first part of the book, you discuss how the image of the sabbath was imposed from the perspective of a conspiracy, a multifaceted conspiracy. The year before however, one of your best friends, Adriano Sofri, had been arrested by the Italian police, accused of murder. Shortly afterwards in The Judge and the Historian, you showed that the circumstances in which he was arrested were extremely suspicious.

Carlo Ginzburg: In actual fact, these two events are completely separate: I think I was already correcting the first set of proofs of the book when Sofri was arrested, and the idea that I had developed about these conspiracies went back a long way. I showed how in France at the beginning of the fourteenth century accusations were made against lepers, Jews and Muslims: they were supposed to have poisoned wells, spread leprosy, and engineered the death of the French people. However, their supposed conspiracy masked another which was all too real, the conspiracy developed by their persecutors who – extraordinarily quickly – had issued the accusations, fabricated evidence, and killed the defendants – and this was going on in many different places. Then the conspiracy changed: the plague became the main threat that everyone feared, and this time the Jews and then the witches were accused of causing the plague. It is particularly interesting that when the witches confessed, often under torture, the inquisitors discovered not only the information they were feeding to them, but that there was something else: transformations into animals and flights to nocturnal gatherings. It was this part which was left over that I discussed in the two following parts of my book and showed how it belonged to a much deeper layer of age-old beliefs.

Having said that, the first part of the book certainly did have a biographical slant, but it was actually the biography of my generation. As was the case for many Italians, terrorism was the defining factor. It all started with the bombs in 1969 that people tried to blame on the anarchists, when in fact they were the result of collaboration between far-right neo-Nazi splinter groups and the secret services. Then there was the terrorist activity on the far left which led to a murky affair: the kidnapping and assassination of Aldo Moro. We dealt with these issues in many different ways, and many books were written, but historians did not examine the issues from the point of view of a conspiracy, despite their long-term impact on Italian politics and wider society.

My argument is that a conspiracy is a caricature of politics. We always tend to dismiss the idea of a conspiracy, thinking that it would be paranoid to consider such an idea. There is a serious aspect, however, which the idea of caricature allows us to understand more easily. That is essentially what I was saying in Ecstasies. I was challenging the idea of a conspiracy being coordinated from a single central point, and, equally, challenging the idea that everyone involved was necessarily acting in bad faith: that would be a caricature. I also provided a more flexible definition of a conspiracy: a series of calculated, coordinated actions intended to steer a series of pre-existing pressures in a pre-determined direction. This definition could also have been used to describe the Years of Lead in Italy.

La Vie des Idées: So it was simply a coincidence.

Carlo Ginzburg : Not exactly. The present brought questions to bear on the past, and vice-versa. As far as the case of Adriano Sofri was concerned, he refused to see the accusation that had been made against him as a conspiracy – at least in public. There were certainly unanswered questions however: the fact that the police did not respond to the preliminary questions of the prosecutor, thereby acting outside any legal framework, the fact that the judges changed tack inexplicably, etc. Conspiracy, in the full sense of the word, is too strong a term to describe what happened. Manipulation, on the other hand, clearly did occur. I pointed out some ways in which events were manipulated in my book, but only very cautiously, and I did not go as far as I would have liked. I had to work within the constraints and ensure that everything was secondary to my main goal: ensuring that Sofri’s innocence was recognised. Everything I said I had to be able to prove. The only approach I could take was one of negation. I therefore showed that the evidence provided was inadmissible, or rather, I tried to shift the onus probandi, the burden of proof. But it wasn’t enough.

Sometimes, whatever happens, history catches up with you. Today, post-9/11, it would certainly be interesting to re-read the part of Ecstasies where I talk about the idea of a conspiracy. In the case of 9/11, too, there was not just one conspiracy: the initial plot, which was absolutely real, gave rise to others which were completely fictitious. A real conspiracy is often masked by a bogus conspiracy: the accusation made against lepers, Jews and Muslims masks the process of constructing the persecution.

How it all began

La Vie des Idées: Let’s carry on talking about Ecstasies. In the two other sections of the book, you examine the popular origins of the witches’ sabbath using very different methods, through morphological comparison. You were faced with these issues of the long-term transmission of knowledge for the first time when, as a young man, you discovered an extraordinary collection of documents on the benandanti. You must have spent almost thirty years investigating!

Carlo Ginzburg: It certainly was fortunate that I came across these documents, but I think that the key factor happened earlier. It is as if I was governed by an invisible constraint that from the outset had made me decide to work not on the persecution of witches and sorcerers, but on the witches and sorcerers themselves. There was clearly an element of paradox as the only archives available – the legal archives – had to be read indirectly, between the lines. Unless – but at the time I wasn’t able to imagine it – a file existed in which the sorcerers and the inquisitors did not manage to communicate, in which the initial misunderstanding remained. This was what happened in a series of trials in Friuli between 1580 and 1650 which were brought against the benandanti, good witches who waged battles in their dreams to ensure the crops were successful. I discovered the records of the trials in the archiepiscopal archives at Udine.

In some ways, you could also say that the direction of my research, which went against the grain of the archives, gave rise to another of my preoccupations, the historical method, which I have never stopped writing about. To succeed in a project like this, I had to think about a historian’s craft. It’s true that in some ways I had two periods: one on witches and another on issues associated with the historical method. The fact remains however that there is a very close link between the two, so much so that it is difficult for me to separate them out.

La Vie des Idées: In fact your use of sources in The Night Battles was already masterly.

Carlo Ginzburg: Perhaps, but at the time this method was only implicit. It was only gradually that I came to understand the implications of what I had done. I had not realised just how exceptional the documents on the benandanti were. Gradually I discovered that there is sometimes a complex relationship between the historian and the inquisitor: emotionally I identified with the defendants, and intellectually I was disturbingly close to the inquisitors. I had a clear conscience when I wrote The Night Battles, but that soon disappeared when I realised how uncomfortably close I was to the inquisitors. I also wrote an article which has not been published in French, The Inquisitor as Anthropologist, where I highlighted these issues.

Similarities and transmission

La Vie des Idées: The other significant moment was when you had just finished writing The Night Battles and discovered the case of a belief that was very similar to the benandanti, but occurred decades later at the other end of Europe, in Livonia. This was the first time that you used the process of morphological comparison, and you went on to use this method throughout Ecstasies.

Carlo Ginzburg: You’re right, it all happened completely by chance. There had been a misunderstanding: I had said to Einaudi that I still had a few changes I wanted to make and my editor thought that I was going to send him a new manuscript! So my manuscript was put to one side for a few months and it was during this period that I discovered this benevolent werewolf who claimed to be one of the “hounds of God”, waging battles against evil witches to ensure good harvests. It was disconcerting because the case raised questions that went beyond Friuli. Due to the delay, I was able to include this example from the Baltic region in my book, but I was only able to develop hypotheses on the existence of a group of beliefs, with very specific characteristics, that were circulating both in Slavic countries and in the Friuli region. In order to properly support this comparison, it was necessary to re-think the relationship between morphology and history.

La Vie des Idées: So your investigation began with the discovery of the Livonian werewolf in 1964 and ended with the publication of Ecstasies in 1989, but surely there was a link in between. In 1976, The Cheese and the Worms was published in Italian. This book was also based on an Inquisition trial, and set out to reconstruct the cosmos of a sixteenth-century miller. You show how the miller appropriates the learned culture of the period in his own way. But you compare the cheese and the worms, which are the origin of the world for Menocchio, with a legend from ancient India, a Vedic legend, merely on the basis of a formal similarity!

Carlo Ginzburg: It is true that I don’t support this point. And what is more, I was wrong. I wrote a new preface, for a Portuguese version of the book, in which I put my research in its context and returned to that mistake. At the same time, I was already thinking about my plans for Ecstasies concerning the transmission of aspects of shamanistic culture in the very long term.

However, I think that if there is a link at all, it is to be found elsewhere, in The Enigma of Piero: Piero della Francesca [1981], where I tackled the relationship between morphology and history head on. For this particular book my research involved interweaving stylistic data and non-stylistic information that was purely historical, such as the patrons who commissioned work, etc. It was only by putting together the two sets of data that I was able to suggest a different chronology for the works of Piero. And there is also a discussion of morphology in my 1979 article Clues: Roots of an Evidential Paradigm. This article is very important to me; you could say that it marked me in a sense. Looking back, it is striking that I don’t talk about evidence in the article, yet this theme subsequently played a dominant role in my work.

On the importance of reading

La Vie des Idées: Even so, in putting forward that hypothesis without supporting evidence, it could be said that you had made an aesthetic or epistemological choice. The similarity could have been explained in different ways but you chose to explain it from the complex perspective of transmission.

Carlo Ginzburg: Let’s try approaching this from another angle. I had been blown away by the sheer novelty of the ideas put forward by Menocchio, but I was playing devil’s advocate in my approach. I tried to demonstrate that everything he had said during his two trials for heresy in 1584 and 1599 came from the books he had read or knew about indirectly. I discovered that there was a gap between what he said to his interrogators and the books that he had read – this point is absolutely key to the book. Menocchio, despite his eccentricities, is always seen as a rich example because there was a historical dimension to his reading. It was at roughly the same time that Roger Chartier, for example, was doing his research on reading, and working along similar lines.

So you have a firm base, and hypotheses. But not everything is attached to the base: I found myself faced with something left over – the cheese and the worms which, according to Menocchio, explain the circumstances of the Creation and which I had not found elsewhere. After writing about the benandanti, I felt as if I was indebted to this collection of documents. This feeling of debt meant that I needed to take up the challenge – the possibility that the beliefs had been passed on over the long term – and carry it through. It was because of this feeling that I worked for such a long time on the project which produced Ecstasies.

There was a time when I thought I would have failed spectacularly if I didn’t manage to write the book. But at the same time, the more I worked on the project, the crazier it became. It was either going to be a small-scale success or a failure on a large-scale: admittedly I was probably more concerned about the second possibility than the other. And then, it was not really me, somehow, that was driving the research. I was guided by the documents that I found, by my hypotheses. At least, I never accepted the boundaries between disciplines that I could have used against myself, saying that I had gone too far, that I was no longer writing history.

La Vie des Idées: Ultimately that is exactly what you have done: you have proved long continuities. This approach is one of the hallmarks of your work.

Carlo Ginzburg: That is true, yet at the same time I am also fascinated by narrow contexts, as was the case with Menocchio. You have the two different approaches, but it depends on the subject. In any extract one always finds either immediate contexts or very distant contexts. The continuity has to be proved, it’s not a starting premise. I am currently involved in two dialogues with the same man, Claude Levi-Strauss. For him, generalising is never the end of a process, it’s a premise; whereas the way I see it, you need to prove the generalisation.


La Vie des Idées: In the introduction to Clues, Myths, and the Historical Method, there is a rather cryptic passage in which you return to the benandanti and talk about your reasons for devoting so much time to them. You wrote that there were other more personal reasons that pushed you in that direction but only appeared to you clearly much later on.

Carlo Ginzburg: I came back to that point slightly later, in 1993, in an article called Witches and Shamans that was published in the New Left Review. I was in Tokyo, where Ecstasies had been translated into Japanese, and perhaps it was this change of scene that helped me to look at the path I was taking more objectively. I had tried to explain how I had started writing this book about the sabbath, making a comparison with the shamans. And I described how one day, quite some time after the Night Battles was published, one of my friends, Paolo Fossati, an art historian, told me just like that, in the street, “When it comes down to it, it’s not really surprising that a Jew should write about heretics and witches”. You could say that it suddenly hit me: it was just so obvious, it was all there right in front of my eyes and I hadn’t realised. Just like everyone else, like millions of other people, I was living proof of the mechanisms that Freud had described: it was important, so therefore it was hidden.

The question was why I knew immediately I had discovered something significant when I found the first document on the benandanti. It was an incredible feeling. I walked out of the archives in Venice, smoking compulsively as I went. Why? Sometime after the event, I told myself that other historians could have written off what they’d seen as a quirk deserving nothing more than a footnote reference. Did this moment sow the seed for the idea that I developed twenty-five years later, in the first part of Ecstasies, this idea that the witches represented the Jews? I don’t think that was the case. Anyhow, even if there was something driving me that was affecting me subconsciously, that something was not simply a matter of me identifying with these people, it was far more complex than that.

La Vie des Idées: In short, it seems that there are two levels to what happened. There was certainly the initial moment when you were transfixed by this document for reasons that you were unaware of at the time. But to pursue this line of investigation over such a long period requires a lot of energy: you owed something to these people.

Carlo Ginzburg: There is this personal side to it all that went a long way: at one point I even had the absurd idea that I had been born to make the story of the benandanti public. That was my justification for it all. I felt that I needed to pay back my debt. The limitations of my discipline could not have prevented me from paying it back. I didn’t want to give up, I saw the challenge through to the end. But it wasn’t as if I started from nothing, I did have models to work from, such as The Royal Touch by Bloch, which really converted me to history.

La Vie des Idées: But perhaps there is a third level, the process of divulging information. In 1986, you talked about “more personal reasons”. In the preface to History, Rhetoric and Proof in 1998, you wrote: “I am a Jew who was born and grew up in a Catholic country; I never had a religious education; my Jewish identity is in large measure the result of persecution.” That’s not exactly the same thing.

Carlo Ginzburg: There are two sides to it. The information was there, but I needed some time to understand its implications. The distance – teaching in Los Angeles for example – helped: I was able to look at Italy and my education from a greater perspective. The other side is the relationship that you have with your moi haïssable (hateful self) as described by Pascal, and I agree with him: it shouldn’t be revealed. Yet if Pascal is right in his way of looking at things, we still need a self to act and think. After all, it’s the self of each individual that gives us access to scientific or moral truth. The self is therefore an imperfect yet necessary instrument. It’s a means, but not an end in itself. Perhaps we are simply less likely to speak about ourselves when we are young. I tried to avoid speaking about myself but clearly, the longer you live, the longer the path that lies behind you. What I wrote there was my contribution to the genre of autobiography. I’m not tempted to give autobiography a try, although it is a fascinating genre.

What we discover and what we leave out

La Vie des Idées: You might not have wanted to write your own autobiography but you are certainly one of the historians who has returned most frequently to the way their intellectual ideas have developed – in that piece of writing in particular and on other occasions too.

Carlo Ginzburg: The idea of sharing things which could help the reader is something I’m particularly keen on. Sometimes these are personal things, but not always, far from it: usually I share reading experiences, what I’ve read, and the way in which I’ve read. Again, I could mention the precedent established by Marc Bloch here. I had been very taken with the way in which he presented not only his results, but also showed how he had obtained them. French Rural History: An Essay on its Basic Characteristics immediately impressed me from this point of view. In a sense, the journey involved in getting there is as important as the results themselves. You could describe it as a pedagogical strategy, even though I don’t really like that word. I have always been drawn to share research in progress, I’ve been able to set out provisional results and provide a rough outline. And this is a very useful process: we need to learn from our mistakes.

Yet behind this issue there is another problem that fascinates me: the relationship between chance and our initial instincts or the provisional judgements we make. We don’t speak enough about this relationship, even though it lies at the very heart of research, because chance is not well regarded and is considered to be incompatible with a scientific approach. Chance is filtered and re-worked by this initial input. Or rather, to put things another way, there is an interaction between chance and what surprises chance: we are not completely in charge of the process. To give you an example: I started working on witchcraft with the rough hypothesis that witchcraft was a primary form of the class struggle. Strangely enough, I found the case that corresponded to the provisional judgement I had made. I discuss this case in the very first article that I published, Witchcraft and Popular Piety. It wasn’t a particularly pleasant experience from a subjective point of view however: I had confirmed my hypothesis but I was disappointed. The end of the article is surprising however. I stated in a rather peremptory way that the remarkable case of Chiara Signorini could be a paradigm case in that it provides a clear example of what happens during interrogations of this sort: the interaction between the judges and the defendants does in fact allow the expression of popular beliefs to come through. The proof only came afterwards, with the benandanti. And so on and so forth. We are also guided by the documents that we find.

La Vie des Idées: We’re still quite a way from psychoanalysis!

Carlo Ginzburg: We certainly are, but then I never agreed to subject myself to the process. Actually, that reminds me of something. If you take the fact that I’m not keen on autobiography and not keen on the teleological aspect of the exercise either, together with these stage-by-stage texts where I described how I got to my results, you could say that what I want to project of myself is more or less what I tried to discover about Menocchio: essentially, what I have read and how. Probably, just as was the case with Menocchio, there will be something left over, something remaining, something that you won’t be able to find in the material I’ve read. But surely there is always something remaining in all of us.

Interview by Florent Brayard on 15 January 2010 first published in and translated by Rebecca Atkinson with the support of Foundation Maison des Sciences de l’Homme.

Dossier's Articles

by Florent Brayard, 15 December 2010

Further reading

 The Night Battles: Witchcraft and Agrarian Cults in the Sixteenth and Seventeenth Centuries, The Johns Hopkins University Press, Baltimore, 1983 [First published in Italian in 1966].

 The Cheese and the Worms: The Cosmos of a Sixteenth Century Miller, The Johns Hopkins University Press, Baltimore, 1980 [1976].

 The Enigma of Piero: Piero della Francesca, London: Verso, 1985 (revised edition 2000) [1981].

 Clues, Myths, and the Historical Method, The Johns Hopkins University Press, Baltimore, 1989 [1986].

 Ecstasies: Deciphering the Witches’ Sabbath, New York: Pantheon Books, 1991 [1989].

 The Judge and the Historian: Marginal Notes on a Late-Twentieth-century Miscarriage of Justice, London: Verso, 1999 [1991].

 Wooden Eyes: Nine Reflections on distance, New York: Columbia University Press, 2001 [1998].

 Le Fil et les Traces (Connections and Clues), Verdier, 2010 [2006].

To quote this article :

Florent Brayard, « Invisible constraints. Interview with Carlo Ginzburg », Books and Ideas , 15 December 2010. ISSN : 2105-3030. URL :

Nota Bene:

If you want to discuss this essay further, you can send a proposal to the editorial team (redaction at We will get back to you as soon as possible.

Our partners

© - Any replication forbidden without the explicit consent of the editors. - Mentions légales - webdesign : Abel Poucet