I'm very grateful to Euan Cameron for such a thoughtful and constructive review, which has forced me to think very hard about some of the broader implications of my own arguments. While it is certainly flattering to be portrayed as a highly original i nterpreter, I had not seen myself as cutting quite such a swathe through the conventional wisdom, nor as so ferocious a critic of others. Clearly my central arguments don't fit at all well with the idea that persecution was simply an epiphenomenon of the 'Godly state'. It seemed necessary to make this plain, while in addition this interpretation looks to be at odds with the facts even on its own terms - I'm relieved to see that Euan accepts this point. On the other hand, I think that there are many val id ways of approaching the material, which would not necessarily clash with my presentation at all. I don't claim that my particular mode of analysis is the only possible one, merely that it must form a major part of any plausible overall interpretation.
One of my prime concerns was to get away from what I saw as an excessive emphasis on formal legal action, and to show how the sources make sense in rather different contexts. This may well have led me to neglect some crucial issues with respect to the legal side of things, although I'd be grateful to be told just what these are and why they matter. Probably because I concentrate on the evidence from witness statements and first interrogations (when the accused were usually still denying their guilt), the similarities looked much more striking than the differences. On the whole I'd still stick to the contention that legal systems usually matter less than how they are used and who is using them, at least in early modern Europe. There are certainly so me problems worth further thought here; a comparison between my findings for Lorraine and those of Walter Rummel and Eva Labouvie for the adjacent regions of Trier and the Saarland suggests some promising lines of investigation.
Where Scot is concerned I plead guilty to a rather sloppy formulation, which made things sound simpler than they really are. On the other hand, I'm not sure that the central point I wanted to make is affected by Euan's very proper rebuke. Scot does i ndeed regard all witchcraft - including the white variety - as fraud or delusion, and uses the term 'witch' for magical practitioners as well as for the old women accused by their neighbours. Yet to me neither his discussion nor his examples - including the Ferrall case - suggest that he lumped all witches together in other respects. Like several other English commentators, he emphasizes the way in which the cunning folk took a leading part in creating and intensifying suspicions. Witches they may be, but at the same time they are surely among those 'witch-mongers' he so bitterly opposes. There is no obvious reference in his book to the possibility that cunning folk might accuse their rivals of black witchcraft, which was the specific point at issue i n my discussion. I might incidentally mention the fascinating and as yet unpublished work which David Wootton has done on Scot, from which it emerges that he may well have been a Familist, and that his attitude to poverty, with its social and economic co ncomitants, was complex and unusual. This should have had implications for his views about witchcraft, with their strong social content.
As just said, other English observers, like Gifford, Perkins and Ady, attached great importance to the role of the cunning folk, whom they often suggested were more culpable than the supposed 'black' witches. This is what I had in mind, along with evi dence from elsewhere in Europe, when I stated that black and white witches were indissoluble parts of the same belief system. Indeed, virtually all serious investigations of witchcraft would lead one to think that this is a normal expectation. If I am w rong on the specific point of the healers' own vulnerability, and they were more likely to be accused than current evidence suggests, I'm not sure how far this alters the general picture. In Lorraine, where I believe them to have been more at risk than i n most parts of Europe, they still make up only a small minority of the accused. Even if one takes some very marginal candidates into account, 10% of those charged with witchcraft is the highest figure that could be sustained for the duchy, from several hundred cases where one has exceptionally full documentation. On the other hand, there is every reason to emphasize the ambiguity of their position; they were playing a dangerous game, which could easily rebound on them. I don't think this is a case of the historian trying to have his cake and eat it, but of a very complex situation which it is hard to describe accurately.
The remark that the specialist witch-hunters were marginalized seems to be more of a comment than a criticism. I would still want to reply that it is actually very difficult to write about them in the sort of detail required if one is to get beyond ge neralities. This is essentially because of the limitations of the source material; as I remark at one point in the book, it is hard to get far inside their minds without direct evidence from their own hands. It's not that they didn't write books or pamp hlets on occasion, just that these are usually rather unrevealing, reporting the supposed actions of the witches in thoroughly conventional terms. Even where there is quite a lot of suggestive material, as for the Trier persecution for example, the loss of the primary trial documents is a serious handicap. The ferocious persecution in the lands of the Elector of Cologne remains particularly shadowy, despite some tantalizing hints about what was going on. Remy is rather a different case; most of his boo k is taken up with reporting the same type of material as is found in the Lorraine archives, so I preferred to rely on these much richer accounts. In truth he doesn't qualify as a witch-hunter in any very obvious sense, despite his stated enthusiasm for the trials. These were already well under way when he became procureur general of the duchy, and there are only a couple of marginal cases where he is known to have done anything to encourage them directly, without obvious success.
Euan Cameron's very reasonable suggestion that people may have been taught to fear witches would presumably require him to assign a greater role to the witch-hunters. It is true that there are a few instances from the fifteenth century where we either know that such people were operating, or may plausibly intuit that they were. Unfortunately there is then a huge gap, stretching virtually from the Malleus Maleficarum in the 1480s to the Trier persecution a century later, when they seem conspic uous by their absence. Massive record losses may be misleading us, of course, but on the evidence I think they are just too rare and scattered to have been effective vectors of new beliefs. In addition, they only reappear when persecution was already ve ry active in many areas. To say this is not to dismiss the idea of 'learned' behaviour out of hand. On the contrary, I'm sure that as the persecution developed it exacerbated fears and generalized beliefs. One of my arguments is precisely that in so do ing it ultimately became a cause of elite scepticism, at a point when growing popular hostility to witches might otherwise have had appalling consequences. There is a crucial interface between official views and popular ones here, whose operation is boun d to remain hidden from us much of the time.
Perhaps the most difficult of all the major issues raised by Euan is how far one can justifiably assume that witchcraft beliefs were widespread in Europe before the legal persecution started. I have always been thoroughly uneasy on this question mysel f, because there are so many problems with the evidence, while the slipperiness of witchcraft as a concept also comes into play. One point I might have made rather more of, in retrospect, is the determined materialism evident in the early modern trials. Wherever possible witches were supposed to have operated through physical means, most commonly the diabolical powder and other poisons; in England the animal familiars usually performed the evil on instructions from the witches. In this sense magic, sor cery and witchcraft all blend into one another, with no clear frontiers. It is certainly possible that the various Anglo-Saxon and Carolingian references to witchcraft are primarily concerned with what anthropologists would call sorcery, yet this wouldn' t affect my argument very much. Europeans don't seem to have supposed, like the Azande, that witches might damage others unknowingly - the deliberate intention to harm was central. Spells, charms, curses, and image magic were all techniques which suppos edly gave effect to malevolence, and we know from the findings at Roman sites and numerous medieval references that many people thought it worthwhile trying to harm their enemies in such ways. My social and psychological explanations don't require that s ocieties should imagine a specific form of diabolical witchcraft, just that they should share a widespread belief in occult powers and secret enemies who used them.
The blunt fact is that we know very little of a systematic kind about popular beliefs before the late middle ages, so that it is hard to see how a convincing picture could be established. We have to rely very largely on the evidence from clerical sour ces, whose authors had their own agenda, and whose very terminology is often ambiguous. On balance I incline towards the supposition that there was quite a well-developed set of witchcraft beliefs current, while accepting that there may have been signifi cant regional variations. What influences me most in this direction is the way that very early trials already contain material which could just as easily have come from the seventeenth century. If John Nider was being truthful about his sources, there w ere cases of this type from the Simmenthal around 1370-1400 (very well analysed by Arno Borst). Kieckhefer's discussion of pre-1500 trials points strongly in the same direction; wherever witness depositions survive, they are full of maleficium accusation s in standard form. Paravy's recent book on the Dauphine suggests a similar pattern there, starting in the 1430s. If we are to suppose that peasants were taught how to make such charges, then there has to be some plausible mechanism whereby this instruc tion took place. This I find hard to identify; the clerical and lay judges who investigated the early cases already seem to be operating on a rather different agenda, so that the dichotomy between popular and elite conceptions of witchcraft is present fr om the start.
One of the most puzzling features of these beliefs is why they seem to be so variable, even capricious, in their operation. This is true at several levels, from deciding whether an illness should be blamed on witchcraft up to the large differences in regional and national experience. Why does one village produce a whole clutch of trials, when nearby communities with apparently identical social, economic, and religious structures have none? Euan suggests this would be easier to explain on the theory that witchcraft beliefs were an acquired cast of mind; I agree in part, as will soon emerge, but still find some serious difficulties with this idea. Early modern villages may have been claustrophobic, but they were not truly isolated, and the trials the mselves show that stories and reputations were transmitted quite freely. Priests and ministers, whom Euan sees as trying to demonize popular magic, mostly appear in the records taking two diametrically opposed positions. Some co-operated enthusiasticall y in the search for witches, often using techniques of divination which hardly differed from those of the cunning folk. Many others were very sceptical, seeing only popular superstition and credulity working to destroy innocent parishioners. Neither att itude really makes them credible as the creators of a belief system; in one they are too dependent on popular notions, in the other too dismissive.
None of this is to say that Euan is wholly wrong; in fact I agree with much of his position, in a revised form. It would be very stupid and wholly unhistorical to think that beliefs just continue unchanged, at a more or less constant level. The troub le is, of course, that we can only really speculate about all this, when the evidence is so defective. The scraps that do survive make me think that some basic notions about malevolence and the infliction of harm on others were widespread in medieval Eur ope. It must have been very important, however, that these had relatively little support from the legal system or church doctrine. When people had little chance of acting out their fantasies or suppositions on a public stage these would have tended to r emain private and shadowy. In addition, if Valerie Flint and others are even partially right about the prevalence of magical practitioners in the early medieval period, and the range of explanations they offered, then this might have marginalized the ide a of witchcraft by neighbours. It would have been only a small part of a vast edifice of supernatural explanations, much of which, as Euan says, laid the blame on demons and other non-human agencies (as with the Anglo-Saxon notion of 'elfshot'). Magical healers would have had a number of good reasons for preferring non-witchcraft diagnoses; these gave them more scope for their role as intermediaries, were more inherently flexible, and avoided alienating potentially dangerous enemies. Indeed many early modern cunning folk still preferred to identify sicknesses with saints who must be propitiated.
This hypothesis - and I claim no more for it - might be extended to argue that as the church gradually succeeded, between the eleventh and the fifteenth centuries, in imposing a more severely Christian vision of the world, and in diminishing the status of magical healers, so it indirectly opened up a larger space for witchcraft. By means we will never be able to reconstruct in any detail, but which probably involved at least a three-way interaction between clerics, cunning-folk, and the people, there came into being a greater public willingness to blame sickness and misfortune on neighbours. As I see it, this would have taken shape first as a discourse, establishing itself at the level of storytelling and gossip. While it is unlikely to have been co mpletely new, it may have become much more central than it had previously been. Such a step would already have been very important, helping to turn the latent fantasies of individuals into shared convictions. Once trials began these added a powerful new element, moving witchcraft out into public view, and legitimating open discussion. These were the means through which people learned about witchcraft, without being 'taught' in any formal sense; they were particularly effective because they allowed a co nstant interplay between official and popular ideas, and because they tapped into such basic psychological structures.
It is inherently implausible that such processes should occur simultaneously across Europe, so there is no reason why beliefs should not have been much weaker in some parts of the Alps than others. Nevertheless, I am uneasy about using the absence of witchcraft from the findings of Inquisitors looking for heresy to prove anything. Investigations of this kind so often came up against a wall of silence that they have to be treated with great caution. As for Salazar, his comment must be understood in c ontext. The persecutions in French and Spanish Navarre have left a good deal of detailed evidence, from which there emerges an extensive local folklore about witchcraft, around which the charges were plainly built. On the other hand, a very high proport ion of both accusers and accused were children, and the great wave of panic on the Spanish side was fuelled by a preaching campaign and the actions of the Inquisitors. Salazar's remark was therefore perfectly valid as applied to most of the cases, but do es not mean that there was no pre-existing structure of belief (and suspects) - Henningsen's meticulous study reaches precisely the opposite conclusion.
It will be seen that although I accept Euan Cameron's very neat summary of my views about the psychological bases for witchcraft accusations, and believe that the cross-cultural evidence strongly supports this picture, I don't want to push the continui ty point too hard. If I avoided a full discussion in Witches & Neighbours, that was partly because I don't think it matters greatly for my analysis of sixteenth and seventeenth century cases. There the evidence seems to me to stand on its own, and to be compatible with different views of how such structures came into being. Moreover, I really do want to insist the need for multi-causal approaches, which avoid giving artifical prominence to any single theme. One point worth stressing is that I believe the patchiness of the persecution has a great deal to do with the powerful forces working to restrain it at various levels. Where these were rendered less effective, as by village committees, complicit local officials, or rulers like the Elector of Colog ne, then persecution could spiral upwards in terrifying fashion. Across most of Europe, however, villagers remained hesitant despite legal systems which seemed to encourage then to take action. Here one needs to look at the ways suspects could defend th emselves, through family networks, threats of revenge, and ultimately the possibility of denouncing others as accomplices. One also has to try and comprehend the cast of mind which could lead someone to denounce a neighbour as a witch, then later make he r a proposal of marriage. Fantasies are full of such ambivalences, which is why they can't bear too much explanatory weight, and the psychological approach on its own won't get us very far in understanding this supremely complex subject.