Witchcraft

Perry Anderson

  • Storia Notturna: Una Decifrazione del Sabba by Carlo Ginzburg
    Einaudi, 320 pp, lire 45,000.00, August 1989, ISBN 88 06 11509 X

Carlo Ginzburg has many claims to be considered the outstanding European historian of the generation which came of age in the late Sixties. Certainly few have equalled him in originality, variety and audacity. He made his debut with a spectacular discovery: the first, and still only, documented case of a magical fertility and funerary cult in the countryside of Early Modern Europe, the trances of the Benandanti in Friuli, stumbled upon unawares by the Roman Inquisition. Next, he transformed the genealogy of religious dissimulation in the age of the Reformation, by tracing the origins of Nicodemism – theological doctrines sanctioning public concealment of private faith – to the defeat of the Peasants’ War in Germany and milieux close to Anabaptism, well before the rise of Calvin, whose attacks on Nicodemism coined the term. There followed his vivid portrait of the autodidact Italian miller Menocchio, whose cosmology of spontaneous generation – the world born as cheese and worms – he referred to a subterranean peasant materialism. Changing terrain again, Ginzburg then suggested a new iconographic explanation of Piero della Francesca’s greatest paintings, linking them through an unnoticed Aretine Humanist to the abortive union of the Greek and Roman Churches, and the crusades projected around the fall of Constantinople. The intellectual unity, and novelty, of these different enquiries can best be grasped in the essays that make up the recent collection Myths Emblems Clues. Its centrepieces are two long methodological reflections, the first on the Warburg tradition of art history, and the second on the heuristics of attribution, from ancient divination to modern connoisseurship.

Ginzburg’s new book, Storia Notturna, more than keeps the promise of this record.[*] It is by far his most ambitious work to date. Subtitled ‘A Decipherment of the Sabbath’, it advances a vast, dramatic reinterpretation of the central image of the European witch craze. Far from being simply a phobic invention of the persecutors, confected from fixed stereotypes of heretical diabolism and garbled scraps of rural magic, the witches’ sabbath reflected the deepest mythological structures of popular culture of the age – a network of beliefs and practices rooted in Eurasian shamanism, stretching from Ireland to the Bering Straits, and running back across millennia past the Ancient World to the darkness of Indo-European and Ural-Altaic origins. In a polemical introduction Ginzburg criticises those historians who have concentrated on the authorities and procedures that set European witch-finding in motion, at the expense of research into the beliefs of those persecuted as witches – Trevor-Roper in the first instance, but also Keith Thomas, charged with reductionism and functionalism. Against this tradition Ginzburg sets what he sees as the superior programme of Lévi-Strauss’s Structuralist treatment of myths as symbolic systems, whose hidden meaning is generated by unconscious operations of the human mind – even though Lévi-Strauss’s anthropology has given insufficient weight to historical research proper. By contrast, Ginzburg’s aim is to combine the morphology and history of the Sabbath – its synchronic significations and its diachronic development – in a single, comprehensive reconstruction.

The argument of Storia Notturna is divided into three parts. The first opens dramatically onto a staccato account of the French pogrom of 1321 against lepers and Jews, accused of poisoning wells in a plot against Christendom orchestrated by the Muslim ‘King of Granada’. It then moves to 1348 and the massacre of Jews as agents of a conspiracy spreading the Black Death, which unfolded further east towards the Alps. In each case, confessions of a phantasmagoric iniquity were extorted, under pressure of torture. By 1380 the Inquisition was ferreting out Waldensian heretics on the southern flank of the Alps. Soon afterwards, Ginzburg suggests, the obsessive fears at work in these persecutions of successive marginal groups were condensed and displaced onto the spectre of a new sect practising witchcraft in the Alpine regions. With this, other dread themes surfaced, absent from the earlier confessions. By about 1440, the full nightmare of the Sabbath – diabolism, anthropophagy, animal metamorphoses, supernatural flight, promiscuity – had been incubated in the Christian imagination.

Ginzburg does not pursue the consequences. Breaking off his historical account here, he switches directly to the meaning of what he terms the ‘folkoric nucleus’ of the Sabbath – identified with the motifs of nocturnal flying and animal transmogrification. The second part of Storia Notturna pursues the archaeology of these motifs. It picks out three cultic origins behind the popular beliefs that went into the compound image of the Sabbath: ecstatic experiences (for women) of a night goddess surrounded by animals, and (for men) of a night battle to ensure fertility or prosperity; and ritual processions (of males) masked as animals. Ginzburg tracks each of these across formidable temporal and geographical distances, starting out from Archaic Greece and Gallo-Roman Gaul: the first in Lombardy, Scotland, Sicily, the Rhineland; the second in Latvia, Dalmatia, Hungary, Romania, Finland, Corsica, the Caucasus; the third in Germany, Bulgaria, the Ukraine. Through every kind of exotic variation, however, all betray a common source – the voyage to the dead undertaken in the shaman’s trance. The journey of the living to the land of death, symbolised in such practices over thousands of years, constituted the clandestine core of the Sabbath as it took shape at the end of the Middle Ages.

In the third part of the book, Ginzburg explores possible explanations for the morphological unity of a folklore extending far into Siberia and Turkestan. He starts by suggesting that it could have derived from the nomadic migrations which spilt out of Central Asia in the eighth century BC, bringing the Scythians – an Iranian people – into the Caucasus and the steppes to the north of the Black Sea, where Greek traders and colonists encountered them, absorbing certain shamanistic features of their culture. In the sixth century, Scythian contingents penetrated south, establishing themselves in the Dobrudja, where they ruled over a local Thracian population, subsequently joined by Celtic settlements. Could this Scythian region have been the original scene of a cultural synthesis, fusing mythological elements from all three peoples into a millennial substratum of beliefs and customs, capable of spreading across the continent and surviving in the depths of folk memory from the age of Herodotus to that of Galileo, if not beyond? Does the remarkable similarity of ‘Animal Style’ art, whose decorative forms stretch from China to Scandinavia in a continuum where the Scythian achievement was outstanding, testify to comparable historical connections? After dwelling on the plausibility of these hypotheses, Ginzburg then points to the limitation of all diffusionist explanations, which leave unanswered the question of why external contact between societies should lead to the internal reproduction of the forms of one in another. The problem posed by the persistence over time and space of shamanistic motifs can only be resolved, he concludes, by postulating the existence of general structural characteristics of the human mind.

To demonstrate these, Ginzburg proceeds to examine – in another sudden shift of focus – myths and rites involving lameness. This motif had already been discussed by Lévi-Strauss, who related it to the change of seasons. Rejecting this interpretation, Ginzburg scours (in the first instance) Greek mythology for every manifestation of a deeper category which he dubs ‘asymmetrical de-ambulation’, in which lameness is only one variant, along with the wounded leg, the perforated foot, the vulnerable heel, the missing sandal. Oedipus, Perseus, Jason, Theseus, Dionysus, Achilles, Philoctetes, Empedocles and a host of other figures display this motif – as do Cinderella, the most far-flung of all folk-tales, and the Chinese crane-dance. Its symbolic meaning is a journey to the world of the dead. But if the pervasive recurrence of this motif belongs to a unitary Eurasian mythology, it is anchored in a universal human experience, ‘the self-image of the body’. Asymmetrical de-ambulation is the privileged signifier of contact with death, because all living beings are symmetrical in form, and among them humans are specifically biped. The impairing of the capacity to walk amounts to putting a figurative toe in the waters of extinction. There is thus in the end an ontological foundation for the symbolisation of the voyage beyond human experience, to the world inhabited by the dead. Myths dictate the limits of their own variation because they are constrained by the formal structures of the imagination.

Storia Notturna ends with a brief Conclusion which is in fact more like a coda. Here Ginzburg suggests that if the image of the witches’ sabbath could so effectively fuse clerical obsessions from above and folk myths from below, it was in part because they shared a common fear of conspiracy whose popular form was the belief that those who had recently died were moved by resentment towards those who were still living. Perhaps too, he speculates, there was a psychotropic element in the trances (hallucinogenic rye or mushrooms) which either contributed to or was projected onto the whole complex. However that may be, the myths which flowed into the Sabbath all converged on the notion of a journey to the beyond and back again, of a crossing over to the world of the dead and returning from it. Ginzburg ends by arguing that the permanence of this theme, through hunting, pastoral and agricultural societies alike, may have a simple but fundamental explanation: the voyage to the dead is not just one narrative among others, but the original matrix of all possible narratives. In the cauldrons of Walpurgis Night are concocted the ingredients of every human tale.

By any standards, this is a bravura performance. It is difficult to think of any other historian who combines such polymathic cultural erudition, grasp of textual and visual detail, and high theoretical aim – not to speak of literary skill. The result is a work of vertiginous effect. There can be little doubt of the audience it is destined to win. To do it critical justice, however, may not prove so simple. For with all its extraordinary gifts, Storia Notturna poses a series of difficult problems, concerning the methods it adopts, the conclusions it reaches, the outlook it suggests. It is best to begin with the first of these.

Ginzburg tells us at the outset that the procedure of his book was inspired by a comment of Wittgenstein’s on Frazer’s Golden Bough, to the effect that mythological materials did not need to be set out historically, as Frazer had done (situating them in an evolutionary sequence), but could equally well be presented ‘perspicuously’ – that is, he explained, ‘just by arranging the factual material so that we can easily pass from one part to the other’ and therewith ‘see the connections’. Hence, Wittgenstein went on – here is the motto Ginzburg took for his research – ‘the importance of finding intermediary links’, ‘as one might illustrate the internal relation of a circle to an ellipse by gradually transforming an ellipse into a circle’.

For Ginzburg this was the charter for the kind of morphology he was looking for. A more formalised version of it was to be found, as it happened, in an essay on ‘Polythetic Classification’ by the English anthropologist Rodney Needham, which Ginzburg duly uses. Needham, too, was much impressed with Wittgenstein’s insights, although he relied, rather, on the familiar text from the Philosophical Investigations which describes the notion of a game as indicating no more than a ‘family resemblance’, without any common feature in the set of which it is used, just as ‘the strength of a thread does not reside in the fact that some one thread runs through its whole length, but in the overlapping of many fibres.’ Whereas monothetic classification requires the presence of at least one common trait in the class identified, polythetic classification – Needham argued – merely demands that each member of the set display a large number of the range of relevant traits, and that these traits are displayed in a large number of the members. He illustrated the basic idea with three descent systems, the first exhibiting features p/q/r, the second r/s/t, and the third t/u/v: such was the type of overlap that sufficed for polythetic purposes.

In urging the importance of this kind of classification for the social sciences, Needham nevertheless entered two cautions. It was the example of the natural sciences, where it was first employed, that lent weight to the validity of the method: yet there the findings of bacterial taxonomy suggested that it did not, after all, make much difference, since a monothetic core of common properties appeared in polythetic classes anyway. On the other hand, whereas in nature there are discrete empirical particulars – elements and particles – from which classes can indisputably be built, no such readily isolable units exist in society. To meet this difficulty, Needham went on to postulate certain ‘primary factors of human experience’ as basic elements underlying the possibility of polythetic classification in anthropology. But he warned that the latter remained a vague notion, since it lacked any rigorous definition of what counted as ‘a large number’, either of features or of members. Ginzburg, for his part, rejects Needham’s primary factors of experience, as a conception too close to Jungian archetypes. But he provides no alternative elements of a formally comparable character in lieu of them; and he neglects the problem of defining the acceptable range of a class altogether. The result is a hermeneutic blank cheque, drawn from an uncritical reliance on Wittgenstein.

This is a slender reed to rely on. Generally innocent of the social sciences, Wittgenstein was attracted to just one dabbler in them – Oswald Spengler. It is no accident that in the very note on The Golden Bough used in Storia Notturna the only example Wittgenstein gave of the method he was recommending should have been Spengler – one that Ginzburg tactfully omits. The naivety of Wittgenstein’s morphological recipes is thus what might be expected. Family resemblance: who has not winced at the fatuities of its discovery, by doting uncle or gaga grandparent, in any improbable infant – or for that matter adult – feature? Intermediate links: by the method of gradual changes, circles can be geometrically transformed at will, and not just into ellipses, but into any number of oval shapes – or indeed into hexagons, triangles or squares. As a principle, the procedure licenses endless approximations.

Myths have always formed treacherous ground for morphological analysis, at once apparently hospitable in their formal variations, and actually intractable in their lack of natural segmentation. The kind of structural analysis practised by Lévi-Strauss always depends on a series of analytic fiats which break up their narrative unity into semantic units at the convenience of the analyst, for rearrangement into putatively underlying patterns. In the absence of independent criteria for selection of the features so translated, the results are notoriously contestable: scarcely one of Lévi-Strauss’s exercises has ever enjoyed a general consensus. Seemingly insensible of the methodological objections so often raised against them, Ginzburg’s only real criticism of Lévi-Strauss is that he is insufficiently faithful to himself. Plus royaliste que le roi, he taxes him with a relapse into a humdrum Frazerian interpretation in viewing lameness as a mere seasonal, rather than lethal motif – or, as he grandly puts it, ‘my Frazer has read Wittgenstein’. The consequences are what might be expected. If polythetic classification is to be operationally defensible, it requires a reasoned demarcation of the features bounding the set in question, and an unambiguous specification of the proportion of them that counts for membership of it. Unless each of these – features and proportions – amounts to a dominant cluster, the class will be more or less arbitrary. Ginzburg does not attempt to establish such dominance. Instead, he picks out those elements in successive myths that interest him, and then connects them as so many ‘intermediate links’ in a chain that confers a common meaning on them. His single-minded quest through Classical mythology for asymmetric de-ambulation is a brilliant feat of imagination. But for all its ingenuity, it rests on a series of calculated extrapolations from the narrative contexts of the myths concerned.

The story of Oedipus is clearly centred on sexual and familial issues. Sophocles paid small attention to his lameness, as Ginzburg himself admits. But he discounts the evidence of Athenian tragedy with the ‘supposition’ that ‘in the oldest version of the myth of Oedipus’, pierced feet were the first stage in a tale concerned with travel to the world of the dead. The portrait of Achilles in Homer does not even mention his heel – which is a late accretion first registered under the Roman Empire. Undeterred, Ginzburg tells us that behind the hero depicted by Homer, and unknown to him, a prior god of the dead ‘has been discerned’, of Scythian stock. The figure of Theseus, pervasively connected to the sea, exhibited no physical defect at all. Ginzburg nonetheless enlists him for his construction on the grounds that he lifted a rock to find his father’s sword and pair of sandals – a case of the circle cartwheeling from triangle to square indeed, since there is not even asymmetry in the de-ambulation here. The legend of Prometheus is in-contesfably focused on issues of knowledge. There is not a hint of a hobble in the ‘fore-thinker’, in any variant. Ginzburg, however, adduces first a Caucasian parallel, in which a hero stealing fire sacrifices some part of himself to a rescuing eagle, who restores it, and then an Italian folk-tale in which another hero, without discovering fire, loses a heel to the redeeming bird. From this graphic example of the method of intermediate links – he casually notes that the theme of intelligence plays no role in either fable – he confidently concludes that in all probability it is ‘pure chance’ that Prometheus suffers no asymmetry of de-ambulation in the versions that have come down to us. Classicists might well retort that the only thing lame about the Greek hero is this explanation of him.

In his enthusiasm, Ginzburg dubs such parallelism between Ancient Greece and modern Georgia or Modena with the same epithet as he used to describe the convergence he found between Menocchio’s cosmogony and the Vedanta or Kalmuck mythology: stupefacente. With the ‘perspicuous’ method of ‘arranging the factual material’, Wittgenstein promised, ‘we can easily pass from one part to another.’ Just so. The passages are much too easy, as the morphology cascades swiftly from one myth to another in an accelerating torrent of identifications – the local evidence for which often consists mainly of variations on the formula ‘it has been suggested,’ which recurs with telltale insistence. Eventually, Greek gods and heroes flow – ‘all but inevitably’ – into the great estuary of Cinderella tales. Here Ginzburg singles out as especially significant those versions which include the collection by the heroine of the bones of an animal who has helped her. He then goes on to argue that ‘the most complete version’ of the fable involved the subsequent resuscitation of the dead animal from its bones. But of the over three hundred stories accounted as versions of Cinderella, from across the globe, fewer than 10 per cent include the gathering of bones, and fewer than 1 per cent – just three cases – involve resuscitation. In such defiance of distributional frequency, it is difficult not to see a preconceived conclusion. The resuscitation of an animal from its bones is, of course, a shamanistic representation of the voyage to the dead.

In his influential essay ‘Clues’, which can be taken as a general historical manifesto, Ginzburg argued for an epistemological paradigm attentive to small traces and discrepancies as the signs of hidden truth – a paradigm whose great modern pioneers were Morelli and Freud. The kind of ‘circumstantial’ knowledge to be wrested from minor clues went back to the first hunters peering at hoof-prints in the ground; it was practised by ancient medicine and divination, and inspired jurisprudence and paleontology before acquiring exemplary modern form in connoisseurship and psychoanalysis. Unlike the quantitative, generalising knowledge pioneered by Galilean physics, it sought qualitative individuation of its objects. As such, it was the appropriate paradigm not only for history but for the ensemble of the human sciences, which over time have ever more assumed it as their model. More generally, indeed, we witness, after the unwise pretentions of Marx, the decadence of systematic thought, and the rise of aphoristic thought, associated with Nietzsche – illumination in the fragment. Although this argument recalls the familiar Neo-Kantian division between nomothetic and idiographic disciplines, since Ginzburg, too, appeals to subjective experience for his circumstantial paradigm, the heuristic emphasis it gives to the vestigial and the anomalous sets it apart. But, of course, these can prompt enquiry in any field of the natural or social sciences, just as they never exhaust any. They provide no special model for the latter, which – contrary to Ginzburg’s suggestion – are certainly not closer to the model of ‘diagnostic’ as opposed to ‘anatomical’ medicine, as a glance at the procedures of economics or sociology would show.

What is more striking, however, is that in arguing for the ‘circumstantial paradigm’, Ginzburg does not attempt to discriminate between his illustrations of it. Necromancy and science, empirical lore and speculative fantasy, jostle side by side in his catalogue of the arts of decipherment. Perhaps this is because Ginzburg assumes there has been a historical winnowing of them, leaving only certified candidates for what he calls the ‘elastic rigour’ specific to them. But the oxymoron says enough. Mentioning at one point Sebastiano Timpanaro’s devastating work The Freudian Slip, Ginzburg says he would reverse its judgment: ‘while Timpanaro rejects psychoanalysis because intrinsically close to magic, I try to show that not only psychoanalysis but most of the so-called human sciences draw their inspiration from an epistemology of divination.’ Timpanaro concentrated his attack, conclusively, on the association of ideas as a method of interpreting the lapsus, showing that it was incapable of not disclosing the meaning Freud assigned them. By analogy with that method, Wittgenstein commended approaching myths and rituals as an ‘association of practices’. This would be one way of describing what Ginzburg has done. The price, however, is the same. Just as, at the level of the paradigm, there is no way of identifying what is not a valid form of divination, so at the level of the morphology there is no point at which the associations need ever stop – falsifications never feature.

The end-result is similar too. Freud’s interpretation of slips and dreams insisted – against overwhelming evidence – that the meaning of every diverse case of displacement, substitution or condensation lay in repressed sexual desire. Lévi-Strauss once compared psychoanalysis to shamanism. But his own interpretation of mythologies reproduces the same schema. Beneath the inexhaustible proliferation of myths, with their luxuriance of every kind of local vocabulary, lies one invariable theme: mediation between nature and culture. Propp, who pioneered the morphological analysis of the wonder-tale, producing the 31 functions of its 177 variants, likewise found just a single master-fable underlying them all. In his case, it was the rite of initiation as a journey to the land of death, transmitted from shamanism. Ginzburg has now taken over Propp’s conclusion, and generalised it beyond the wonder-tale, to the farthest-flung corners of Eurasian mythology. The fascination of the data he assembles is beyond question. But once again, what is striking is the contrast between the richness and variety of the materials, and the paucity of the meaning to which they are reduced. Seeing the mote in the folklorist’s eye, the anthropologist complained of the procedure, while reproducing it on a grander scale. The historian, criticising each, goes further, joining fable and myth together in a cosmic one-liner, encompassing all the narratives ever told.

In venturing this final step, Ginzburg moves beyond the formal bounds of his own enquiry. The outline of a unitary Eurasian substratum of beliefs and rituals dissolves into universal categories of the human mind. In fact, the latter part of Storia Notturna continually hesitates between these two. He poses this as a dilemma between two kinds of explanation of the regularities he finds: cultural diffusion (alternatively, common descent) or psychic uniformity. Declining to opt for one or the other, he exploits both possibilities. It is their combination that suggests the ‘interweaving’ of history and morphology at which he aims. But their simultaneous presence in the text looks more like tactical reinsurance than theoretical synthesis, for they cannot logically be reconciled.

On the one hand, there is the hypothesis of a specifically Eurasian mythology based on the practices of shamanism. This line of thought has its own history. In the Russian emigration between the wars, ‘Eurasianism’ was an Orientalising version of the Slavophile tradition, whose leading theorist was N.S. Trubetskoy, the founder of Structuralist phonology. The ‘Presentiments and Subversions’ of the Eurasian Manifesto (Sofia 1921), already preoccupied with unitary folk culture, have an involuntary echo in the ‘Conjectures’ of Storia Notturna. Shamanism, for its part, received heroic exploration from Mircea Eliade, in a study tracing it far from its North and Central Asian homelands, to Germans, Greeks, Scythians and many other peoples. It was Eliade who emphasised the difference between the ecstatic states of shamanism and possession – the shaman controlling communication with the spirits of the dead, the possessed controlled by them. Ginzburg adopts this opposition, and makes of it a cultural frontier. Eurasia is characterized by the ubiquity of shamanism, Sub-Saharan Africa by possession. The Cinderella cycle and scapulimancy, likewise, are common to the former and unknown to the latter. But if this is so, appeal from correspondences between Indo-European and Ural-Altaic ecstatic forms to universal categories of the human mind is ruled out.

On the other hand, Ginzburg states unambiguously that ‘long ago I set out to demonstrate experimentally, from history, that human nature does not exist; 25 years later, I find myself maintaining the exact opposite.’ Although his Eurasian conjectures provide no support for it, the claim itself may of course be independently valid. Certainly there is nothing outlandish about it. His way of construing the notion of human nature, however, is quite particular. Loyal to his Structuralist inspiration, he interprets it in a strictly intellectualist sense. By human nature here is understood the mechanisms of the human mind. Needs and emotions, which might be thought prime candidates, do not figure. Levi-Strauss has announced: ‘I cannot give these turbulent forces primacy; they erupt onto a stage already constructed and patterned by mental constraints.’ Needham, seeking the primary factors of human experience, goes out of his way to deny that affects count among them – no inner turbulence is universal: ‘anger’ in another civilisation is not equivalent to anger in our own.’ What is universal is a common repository of ‘collective representations’, derived from the properties of the cerebral cortex, which can be viewed as close to Jungian archetypes. Ginzburg rejects even these as too concrete, for transcendental operations of the mind whose symbols are not so readily intelligible. But he, too, toys with the idea of fitting Jung with a materialist pediment – that is, giving universal myths a somatic foundation. In the end, the deathly meaning of asymmetric de-ambulation is traced to an exigency of the self-representation of the body. At first sight, the case Ginzburg makes here looks plausible. But it rests on the slimmest of documentary evidence: one possible interpretation of a single myth from the Moluccas is the only empirical referent supporting it. The general argument depends, moreover, on the identification of humanity with the capacity to walk on two legs – a gait that is shared, however, by various animal species (not just apes and bears, mentioned by Ginzburg, but more generally by birds). The human species is more frequently identified with language, and one has only to think of the ease with which the same case could be made for (let us imagine) myths involving ‘irregular articulation’ – muteness, hoarseness, stuttering, lisping, whispering, mumbling and so on – to see how inconclusive it is likely to be. The problem is not the conviction, eminently reasonable, that there is such a thing as human nature, but where to locate and how to define it. Collective representations, however they are conceived, are unlikely to be the main road to an answer. Asymmetric de-ambulation, for all the ramifications it acquires, will not support the anthropological weight put on it.

What then is its more specifically historical relevance to the original theme of the book, the witches’ sabbath? Rather little, it must be said. Many fantastical elements went into this Christian-pagan brew, but one thing it did not include was this. The only connections Ginzburg is able to offer between the ostensible subject and actual climax of the book is one limping boy in a (third-hand) version of a Livonian werewolf’s professions, and a lame Aegean bogey. Neither has any direct relation to the Sabbath. Even by the method of intermediate links, this is a tenuous thread. The first part of the book must be regarded as an effectively independent enquiry, to be judged on its own merits. How are these to be assessed?

Ginzburg’s ‘decipherment of the Sabbath’ involves two principal claims. The first concerns the configuration of the image of the Sabbath itself. Where Trevor-Roper, following a long tradition, viewed this as essentially a theological fabrication incorporating jumbled fragments of peasant gullibility, Ginzburg in effect reverses the order of importance within the mixture, indicating popular belief rather than learned fantasy as the core of the phenomenon. This judgment is never quite spelt out, but it is insistently conveyed by the use of the term ‘nucleus’ to describe the folkloric elements of the Sabbath. The grammatical ambiguity of such pronouncements as ‘here lies the folkloric nucleus of the Sabbath’ is unlikely to be casual: Ginzburg employs the notion to suggest both coherence and centrality. What is his case for them? Here the problem of methodological imprecision returns. For the criteria required to decide what is dominant (and what is subordinate) in a composite formation are missing. So too, in fact, are criteria for defining the composition. For Ginzburg, there are essentially two layers to the Sabbath: satanism on high – the pact with the devil; and shamanism below – the voyage to the dead. The latter is the organic substratum of the phenomenon, whose originally beneficent meaning (the shaman as warrior of the good) is converted into a fearful ceremonial of evil (the witch as servant of Satan) by the artifice of the former.

What this dualism, of sanguine folk myth deformed into vicious élite craze, omits is popular fear of witchcraft itself. General and intense belief in maleficium, the sorcerer’s curse, as the agent of every kind of material injury or misfortune, was a fundamental condition of the witch craze. Massively documented by Keith Thomas and others, it is virtually ignored by Ginzburg. But it has much more obvious claims to be regarded as the folk soil of the phobia than anything to do with ecstatic trances or monosandalism. In Hungary, a region where the presence of a shamanistic element in village life was probably stronger – certainly better documented – than anywhere else in sub-Arctic Europe (the well-known figure of the taltos), it played a discernible role in only about 2 per cent of witchcraft trials. The malefice, on the other hand, was everywhere; even the Benandanti, it has been pointed out, were involved in the transactions of the curse. The paradox of Ginzburg’s account of the Sabbath is that it tends to forget the witches – in the most familiar sense of that term. In sociological terms, this is a central lacuna of his study.

How far, on the other hand, does Ginzburg’s historical narrative sustain his ethnographic view of the Sabbath? Tracing its genesis to the French pogrom of 1321, he stresses more than previous historians the role of deliberate machinations by municipal, royal and noble authorities in fanning a popular hysteria against lepers and Jews from which they stood to benefit economically. ‘Plot’ there was, but on the part of Christian magistrates rather than against them. Here, at the outset, initiative is firmly imputed to those above. But by the time of the Alpine witch-trials of the next century, with which his narrative concludes, Ginsburg goes out of his way to emphasise that in the emergent image of the Sabbath ‘anti-heretical stereotypes’ unleashed from above were ‘only a secondary element’. Norman Cohn is taken to task for exaggerating their importance. The claim is unpersuasive. Cohn’s demonstration of the constancy of the stereotype of the secret sect – practising promiscuity, blasphemy, anthropophagy – from pagan accusations against the Early Christians onwards, loses no relevance with the arrival of the Sabbath, since these were always the capital iniquities ascribed to it. The folkloric themes of nocturnal flight and animal metamorphosis were picturesque embroidery on this background – in themselves moral small change dismissed as no graver than idle fancies by the Canon Episcopi in the Dark Ages. If any elements were secondary in the Sabbath, it would logically and functionally be these. As it happens, it is Cohn who provides the best explanation for a fusion of motifs: the inclusion of nocturnal flying in the anti-heretical stereotype allowed for multiplication of suspects, by magical assembly of a host of secret evil-doers.

Beyond these structural considerations, there are also chronological difficulties in Ginzburg’s reconstruction. His opening account of the pogrom of 1321 is a masterpiece of laconic narration, displaying all his exceptional gifts as a writer. Much of the power of his storytelling, here as in The Cheese and Worms, lies in the understatement of its spare, concise prose. But this very conscious art is also, in another and less obvious way, highly theatrical. Ginzburg’s narratives proceed by short, numbered paragraphs, with minimal connectives between them. The device of enumeration, typically used since Spinoza to suggest rigorous logical deduction, here evokes dramatic sequence: scenes or takes for stage or screen. This a very effective way of unfolding a story. But it has a particular drawback as a method of writing history: it too often depends for its effects on withholding information. Ginzburg’s way of recounting events militates against providing their context, which might slow the speed and diminish the surprise of the tale. We are told very little, in the end, about the Benandanti beyond their fabulous adventures, or about the Friulian society that enclosed them, about the village life that harboured his miller, Menocchio, or the Inquisitors who cross-examined him. So, too, France in 1321 is presented ex abrupto, shorn of socio-political perspective. The lack of any informational frame – general features of France in the early 14th century, immediate situation of the kingdom, policy and personality of Philip V, position of lepers and Jews – quickens the sinister drama, but also deprives it of depth.

One effect of this is especially relevant, and questionable. In representing the French massacre of that year as the starting-point of the fuse leading to the European witch craze, Ginzburg neglects the cause célèbre that preceded it – the extermination of the Templars unleashed by Philip IV in 1307. The charges made, and confessions extracted, in this political repression were in certain respects closer to the phantasmagoria of the Sabbath than the accusations against lepers and Jews. Conforming to the ‘anti-heretical’ rather than ‘folkloric’ nucleus, however, they play no role in Ginzburg’s account – despite the fact, critical for a local aetiology, that the chief leper confession of 1321 borrowed from the show trial of the Templars 13 years earlier.

The more general problem posed by the timing and trajectory of the witch craze, however, remains. What generated the new persecutory hysteria that came to be driven by the image of the witches’ sabbath? He criticises Cohn for focusing on a single long tradition of anti-heretical phobia from the time of Roman emperors to that of Renaissance Popes, at the expense of the ‘evident cultural discontinuity’ introduced by the idea of a sect of witches. But his own concentration on a sub-stratum of shamanistic trance-beliefs enduring across many thousands of years, raises even more sharply the question of what it was that precipitated the emergence out of them of the Sabbath. The explanation he offers is conventional and cursory – little more than generic reference to the socio-economic crisis of the 14th century. But reliable evidence for the crystallisation of the Sabbath starts about a hundred years later: the witch craze reached its height in the relative prosperity of the late 16th century; persisting through the depression of the early 17th century, it eventually subsided, in the main regions infected, before economic recovery was in sight. Moreover, its geographical incidence was very uneven. In Western Europe, England – lacking inquisitorial procedures and addiction to torture – was largely exempt from the obsession with the Sabbath; also relatively unaffected were the United Provinces. In Eastern Europe (where so many of Ginzburg’s examples of shadow shamanism come from), the Orthodox Church paid little attention to it, and major persecution was confined to Catholic regions, particularly Poland. Between these zones, the real epicentres of witch-hunting were Germany, Switzerland and France, where, Brian Levack has estimated, perhaps 75 per cent of all prosecutions occurred. The fundamental enigma of the European witch craze is the pattern of its development in time and space: why it erupted when it did, whom it attacked, why and how it affected certain zones yet passed by others, why and when it petered out. In the answer to these historical questions must lie the key to deciphering the Sabbath. For all the steady gains of later research, the intellectual firmness and clarity with which they are posed in Trevor-Roper’s essay have yet to be equalled.

If Ginzburg’s approach is so different, his aversion to this precedent so marked, the reason lies in the most constant conviction of his work. Tout ce qui est interessant se passe dans l’ ombre, ran the epigraph to The Cheese and the Worms, from Céline. The hidden side of history is where the truth resides. In one of the essays in Myths Emblems Clues, Ginzburg maintains that no cultural opposition is so universal as that between the values ascribed to the positions of High and Low – the former being always equated with what is better, the latter with what is worse. He overlooks another pervasive opposition in our culture, which reverses it: the Superficial and the Deep. Here, what is below inherently excels what is above. But the surface covers the depths, and must be parted to get to the bottom of things.

If there is a single assumption that unifies all of Ginzburg’s versatile work, it is this: that the deeper something lies, the more significant it must be. Nowhere has this belief acquired such writ as in Storia Notturna. The meaning of the Sabbath that matters is to be found at a subterranean level of human imagination stretching back from European fertility or funerary cults, through Greek myths and Scythian ornaments, to Siberian or Mongol shamans, perhaps to Paleolithic bands before them – all revealing symbolic traces of the voyage to the dead. However hazardous some of Ginzburg’s interpretative connections may be, his general argument for the persistence of motifs of shamanistic origin over a very long span of time can readily be accepted. But persistence is in itself no warrant of significance. What is missing from Ginzburg’s account is that erosion of meaning which is such a large part of any cultural history – the familiar process whereby customs or beliefs once centrally active become in altered conditions sporadic or marginal, and then lose their sense altogether as they are surcharged with further developments which incorporate or efface them, and cease to be understood.

Ginzburg himself notes often enough that Classical authors failed to see the significance of asymmetrical de-ambulation in their myths; indeed he complains of their ‘rationalistic’ reconstructions of mysteries that interest him. But he does not draw the obvious conclusion, of a symbolism gone dead. As facts of cultural history, the realistic explanations of the single sandal attempted by Thucydides or Servius are in many ways more interesting than its aboriginal sense – since they obviously tell us more about Greek and Roman society. Shamanism in its Siberian or Lapp homelands was a central institution of rudimentary hunter-gatherer or pastoral tribes. Whatever might have passed from remote social backgrounds, possibly akin to these, down to the Mediterranean city-states would necessarily have assumed a different weight and meaning – still more so, if there were such continuity, in the societies of late Medieval or Early Modern Europe. Ginzburg insists on the parallels between shamanism and ecstatic trances of the kind he found among the Benandanti. Yet all of these were clandestine, whereas – in Eliade’s words – ‘every genuinely shamanistic seance ends as a spectacle unequalled in the world of daily experience’: a fundamental difference. Ginzburg finesses it with a structuralist flourish. A homology nevertheless holds by inversion: if shamans were public, their dreams were individual battles for the good, whereas the oneiric combat of the Benandanti was collective, even if its performance was private.

Whatever the merits of this argument, the salient point it avoids is that the historical function of ecstatic experience had changed completely in the interim – from socially-enacted drama to furtively secreted reverie; and its subjective meaning would have changed too. Similarly, the paraphernalia of broomsticks and toads are more like dim vestiges of a half or wholly mislaid past than living emblems of the ‘primary nucleus’ of the Sabbath. In social life, what is older and in that sense deeper is often more trivial: something that has survived just because it has been reduced to insignificance. In a famous retort to Lévi-Strauss, Jack Goody once remarked that the persistence of certain dishes might be explained not by their symbolic meaning, but their indifference to it, as simple pegs of existential continuity. The same could, of course, be said of many features of dress. This is the general phenomenon of buttons on the cuff. Words and myths, unlike objects of material use, are inescapable signifiers. But they, too, are always liable to de-signification, the process Ginzburg’s morphology consistently forgets. The only difference is that in their case it typically involves the cancellation of one meaning by the superimposition of another, which can easily contradict the first.

To seek to determine the essential meaning of a Classical legend or a Medieval phobia by appeal to a shamanistic origin is not unlike the attempt to demonstrate the contemporary significance of a philosophical concept by recourse to the etymology of its pre-Socratic roots, a procedure favoured by Heidegger. But just as the words of a language can survive and evolve through the obliteration or reversal of their earlier meanings, so may the elements of myth: in neither case does what come first have any semantic privilege. The temptation to equate persistence with significance has its own promptings, however. Appropriately, Wittgenstein’s commentary on Frazer gives revealing expression to one of them. Lamenting the ‘narrowness of spiritual life we find in Frazer’, Wittgenstein deplored his rationalist analysis of magical cult or ritual, whose mystery was better left to reverent contemplation: ‘even the idea of trying to explain the practice seems to me wrong-headed’ – ‘we can only describe and say, human life is like that.’ Above all, he rejected the idea that there had been any basic change, let alone for the better, in human sensibility with the development from magic through religion to science, as Frazer assumed. The simplicity of Wittgenstein’s argument is unmatched: ‘I wish to say: nothing shows our kinship to those savages better than the fact that Frazer has at hand a word as familiar to us as “ghost” or “shade” to describe their views’ – ‘a whole mythology is deposited in our language.’ Here the ideological drift of the fallacy of continuity is ingenuously clear.

Ginzburg tells us that his first and most enduring ambition was to supersede the alternatives of rationalism and irrationalism. Similar hopes are often expressed of the opposition between materialism and idealism, or left and right. It is rarely difficult to see which term of the two pays the cost of the operation. Ginzburg ends his introduction to Storia Notturna with the claim that:

the immensely ancient myths which flowed, for what was after all a short time (three centuries), into the composite stereotype of the Sabbath survived its disappearance. They are still active. The inaccessible experience which humanity has for millennia symbolically expressed in myths, fables, rites, ecstasies, remains one of the hidden centres of our culture, of our mode of being in the world. The attempt to know the past is also a voyage to the world of the dead.

This declaration of faith in the perennity of the nocturnal region of twisted gait and swaying trance is, of its nature, unadorned with evidence. It is doubtful whether Ginzburg would welcome occult magazines and palmistry stores as witnesses. He prefers instead the self-enclosing gesture with which Lévi-Strauss designates his own analysis of myth as itself another myth, engendered by the common substance of the human mind. So the sign of the hidden centre becomes the enterprise of the historian himself – in this place of concealment, mage and scholar are one. There is no need to dwell on the weakness of the conceit in either case. It functions less as argument than as index of an outlook. In Ginzburg’s case, what defines that outlook is the call of the deep.

Paradoxically, however, the very consistency with which he has pursued underground cultural continuities provides its own antidote to them. The Benandanti, peasants practising somnambulist magic to improve the harvest or propitiate the dead, were the bearers of a ‘genuinely popular stratum of beliefs’, ‘with roots stretching far back into time’, to an ‘antiquity when they must once have covered much of Central Europe’. The Nicodemiti, intellectuals justifying religious dissimulation in the name of Christian detachment from outward rites, betokened a ‘deeper and more homogeneous stratum’ of faith beneath ‘the conflicts agitating the surface of European religious life in the 16th century’. The miller Menocchio, village philosopher expounding a materialist cosmogony without magic or Christianity, expressed views and aspirations whose ‘roots were sunk in an obscure, all but unfathomable stratum of remote peasant traditions’. Diana, Cinderella, Cordelia, heroines of myth, fable and drama, spring from a ‘subterranean stratum of unitary Eurasian mythology’, a ‘stupefying dissemination of shamanistic traits’, in which are deposited ‘a history of thousands of years’. Behind Freud diagnosing a disorder or Berenson telegraphing an attribution can be glimpsed ‘perhaps the oldest gesture in the intellectual history of humanity: the hunter crouched in the mud peering at the tracks of his prey’. The vocabulary of depth and diuturnity is unvarying: but its objects are so various that they cancel each other. How probable is it that both ecstatic voyages to the beyond and robust materialist denials of the divine were ancient peasant traditions in the same Friulian hills – existing beneath the surface of a European Christianity whose own divisions were also traversed by a clandestine movement profounder than Catholic or Calvinist confessions? Each of these research programmes has yielded fascinating empirical fruits. But it is a metaphysical predisposition which repeatedly projects their results downwards or backwards. The decipherment of the Sabbath suffers from it. Ironically, perhaps the most unanswerable objection to the thesis that its ‘primary nucleus’ derived from popular ecstatic myths is supplied by Ginzburg’s work itself. For the only investigated case of such cultic trances, the Benandanti, was treated with offhand indifference by the Inquisition. So little alarmed were the witch-finders by the quaint beliefs of the night-walkers that not a single prosecution was ever concluded against them. It is difficult to avoid the conclusion that the crux of the fantasm lay elsewhere.

Ginzburg once wisely remarked that an ideological option may vitiate the documentary findings of an author and yet also be the condition of them. He was speaking of Georges Dumézil, one of the pioneers of comparative mythology, in whom he detected possible pre-war sympathies with Nazism. Mircea Eliade’s Myth of the Eternal Return, another landmark in the field, was born of the defeat of the Iron Guard in Romania, Ginzburg observes elsewhere. What of Ginzburg’s own options? Reflecting on its pathos of human nature, the distinguished Italian critic Franco Fortini has termed Storia Notturna the work of a liberal conservative. This is certainly unjust. Ginzburg’s original inspiration, from which he has never departed, lay rather in a kind of populism, as he himself has noted. But it is true that populism lends itself to a number of timbres, according to its perception of the people.

Ginzburg’s early work reflected the insurgency of the late Sixties, whose mass character was more pronounced in Italy than in any other European society. It spoke of class culture and repressive tolerance, peasant war and social utopia. With the subsidence of the Eighties, the tone has changed. Ginzburg now declares: ‘Growing doubts about the efficacy and outcome of revolutionary and technocratic projects oblige us to rethink the way in which political action is inserted into deep social structures, and its real ability to alter them.’ Storia Notturna is still dedicated, in the spirit of Walter Benjamin, to the history of the vanquished. But now the accent is on the simple perdurance of popular beliefs, across any number of social structures and through every kind of historical and civilisational change: they are anchored in the operations of the human mind. In the process, an immense cultural halo is swung round the gyrations of the shaman, in which we read imprinted as in a Trecento image the glowing letters of Greek Legend and Universal Fiction. Thus are the folk shards of witching hour and animal familiar – Trevor-Roper’s ‘mental rubbish of peasant credulity’ – redeemed as they enter into and yet pass beyond the Sabbath.

The earlier Ginzburg warned against just such idealisation. Commenting on a contemporary Italian cult near Salerno, in which a local woman periodically assumed the personality of a dead nephew, he did not evoke the grand continuity of Eurasian ecstatic traditions, but wrote: ‘In wretched and disintegrated conditions, religion helps men and women to bear a little better a life in itself intolerable. It may not be much, but we have no right to despise it. But precisely because they protect believers from reality rather than prompting or helping them to become aware of and change it, such popular cults are in the end a mystification: to overvalue them in populist fashion is absurd and dangerous.’ It would be good to hear that voice once again. One much-needed word has disappeared from the vocabulary of Storia Notturna: superstition. It is salutary to recall that a Hungarian scholar, describing the poor, sparse societies in which it once dominated, could speak of ‘the misery of shamanism’. That judgment is too strong for what was once a coherent set of beliefs, giving moral shape to a world. But superstitions are the scrambled relics of belief, no longer comprehended: might their role in the history of the Sabbath merit such a verdict? The perspective of post-rationalism – if that’s what, in tune with the times, we should call it – would refuse the thought. Yet what is true of Dumézil and Eliade holds here too. The ideological option may affect the empirical materials, but without it they would not in all probability have come to light. Storia Notturna may overshoot its resources. But these are more than rich enough to justify the enterprise.

[*] Ecstasies, an English translation of Storia Notturna, will be published by Radius next year. The translator is Ray Rosenthal. Myths Emblems Clues, translated by John and Anne Tedeschi, was published by Radius in January (231 pp., £16.95, 0 09 173023 6).