The cult of St. Guinefort, who specialized in dispatching sickly children, astonished the inquisitor who happened upon it in the Dombes, a region north of Lyons, around the year 1250. When he asked about this saint he had never heard of before, he learned to his further dismay that Guinefort was a dog, a greyhound to be precise, who had been killed in the following way.

At a chateau near Neuville there lived a lord, his lady, and their infant son. One day when both parents and even the nurse were away from the chateau, a large snake came in and headed for the infant’s cradle. The lord’s dog followed the snake and attacked it just as it reached the cradle. A ferocious battle ensued in which the cradle was toppled over and the snake, after severely wounding the dog, was killed. The dog dragged the dead snake away and then settled down, all smeared with blood, by the cradle.

The nurse returned, assumed that the dog had killed the infant, and screamed; the lady did likewise; the lord, when he arrived, drew his sword and killed the dog. Only then did they find the infant safe and sound. Moreover, once they discovered the dead snake, they were able to reconstruct the entire scene. With great remorse, they buried the dog in front of the chateau and planted trees about the grave. By divine will, so the story concludes, the chateau crumbled to ruin and the land around reverted to wilderness.

The local peasants often returned to the site to honor the faithful martyr. They called him Guinefort, and when they prayed to him he responded with miracles. His cult was appropriately focused upon the care of children. Women especially took their sickly offspring to this shrine. An old hag revealed to them the secrets of a complicated ritual that included offerings of salt, tossing the naked child back and forth between trees of the sacred grove, calling on fauns to come out of the forest to reclaim the sick child they had previously swapped for the real, healthy child, and asking for the real one back. Children were left alone, exposed on the ground with candles burning dangerously close to their heads, or were dunked nine times in a nearby river. Guinefort did not undertake lengthy cures. Instead he gave speedy answers to anxious mothers: children either expired after this rite or else they were in fact healthy and survived.

With the publication of Le saint lévrier, we have an opportunity to compare two learned investigations of this rustic cult, one by the thirteenth-century inquisitor, Stephen of Bourbon, and now the other by a French historian, Jean-Claude Schmitt. Both were drawn to examine the cult by their professional training and mission. For Stephen it was a dangerous intrusion of pagan superstition into matters where Catholic orthodoxy claimed an exclusive franchise. For Schmitt, the same phenomenon provides a rare glimpse into the popular subculture that the Church had mostly obscured if not actually stamped out. Schmitt is as committed to recovering and studying this subculture as Stephen was to uncovering and suppressing it.

Stephen of Bourbon joined the Order of Preachers (the Dominicans) in Paris in 1217. St. Dominic had just gained approval for his new order and his followers had arrived in Paris in search of recruits. The Dominican program centered on revitalizing the Church’s preaching and penitential ministry; the papacy soon added to this program the responsibility for administering the Inquisition. For four decades Stephen’s home base was the Dominican convent in Lyons, from which he went out to preach, hear confessions, and conduct inquests into heresy. His writings include a handbook for preachers, a new genre, which contained nearly 2,900 exempla—edifying stories or parables for use in sermons. His sources included the Bible and saints’ lives, stories told to him, and personal experience.

The matter of St. Guinefort is set forth in an exemplum based on Stephen’s experience as a confessor in the Dombes. There he encountered several women who confessed that they took their children to St. Guinefort, and so he learned the story we have retold above. About the cult he observed that any child who could survive such treatment by an infanticidal mother “had tough guts” (habebat dura viscera). Then he went to the sacred site, where he called in all the faithful from the countryside around and preached against the cult. He had the dog dug up and the trees chopped down; the bones and logs were burned together in a bonfire. Then he got the local lords to agree to seize and sell off the belongings of anyone who henceforth came to that place to practice the now officially suppressed cult.

Stephen went away apparently satisfied, for only a thorough triumph of religion over superstition would have served as worthy material for a sermon. His introduction to the exemplum explains that idolatry, the granting of divine honors to demons or other creatures, is among the most outrageous forms of superstition.


Stephen’s exempla collection was first published in 1877, but only in the past decade have exempla come under the systematic scrutiny of social historians, in particular historians in the Paris seminar run by Jacques Le Goff in the discipline he calls “Historical Anthropology of the Medieval West.” The main problem for these historians is precisely the relationship between popular and learned cultures, and the principal model available to them has been the essay of 1970 by Le Goff on St. Marcel of Paris and the dragon.1 Le saint lévrier is the first published book to emerge from this seminar.

Schmitt begins by showing the similarities of “structure” between the story of Guinefort and others, some of them earlier. In Stith Thompson’s Motif-Index of Folk Literature he found an entry that reads: “Dog defends master’s child against animal assailant.” The earliest known version is in sixth-century BC Sanskrit, and there are several versions in various Indo-European languages, including one in Old French of 1155 and one in Latin of 1184. But it is no easy matter to get from these precisely dated texts to the illiterate peasants who prayed to St. Guinefort. The oral traditions of the peasantry, which may be rooted in ancient Indo-European culture, are, according to Schmitt, as plausible a route as any direct descent from one of those twelfth-century texts.

This is also not a simple matter of textual transmission because, as Schmitt makes clear, both parts of the story, the saint’s legend and the peasants’ rite, include details—not all found in earlier versions—that correspond with what we know of thirteenth-century society. In the legend, we find the chateau, the lord, the lady, the male heir to the aristocratic lineage, and the greyhound—no common dog but one held on high esteem by a class of warriors-turned-hunters. From Mary McLaughlin’s study of medieval child-rearing we have come to expect to find a nurse in such a household.2 Moreover, any settlement called Neuville was of recent foundation, for the demographic and economic expansion of the twelfth century had fostered the growth of new towns, with names such as Newton, Newport, Villanova, Villeneuve, or Neustadt. And Schmitt, drawing on Pierre Toubert’s social history of central Italy, shows how, still within the age of expansion, the cultivation of marginal lands often went too far and the abandonment of such lands and their reversion to a natural state followed as a matter of course.3

All the parts of the rite had clearly established folkloric meanings at the time as well, from the offerings of salt to the tossing and dunking of children. The belief in changelings holds a prominent place in Schmitt’s analysis. If the sickly children involved were really bound to die soon, then the cult of St. Guinefort, he writes, had less to do with infanticide than with the need for adults to find reassuring ways to confront infant mortality.

The story is therefore much more than an old tale repeated in the thirteenth century; it is full of contemporary details. However the peasants got the story, it has reached us only as refracted through the cultivated mind and Latin vocabulary of Friar Stephen. Schmitt thus gives persuasive testimony against the old notions that culture is transmitted in a vulgarized form from a “higher” class to a “lower” one: he shows that mutual influences were at work as the tale took form.4

But did the peasants make up the name Guinefort? Not even that, for Schmitt found that since the eleventh century there existed across the Alps at Pavia a cult dedicated to Sanctus Guinifortus (man and martyr). By pursuing the name, the feast date, and the emphasis on sick children, Schmitt was able to establish the existence of fifty-nine centers, eight in Italy and the rest in France, where some variant of this cult was observed. And through an ingenious study of the circumvectio, a fund-raising campaign by monks who paraded relics from one village to another, Schmitt offers a convincing explanation of how Guinefort could have arrived in the Dombes. Thus the unusual mixture that took place there consisted mainly of the name of a saint and martyr along with the principal components of a cult brought by monks from Pavia, joined to a nexus of traditional folklore about languishing children and a very ancient legend about a canine martyr to babysitting.

So much for the event and the setting, l’évènement et la conjuncture Braudel would say.5 Had Schmitt gone no further he would have made a major contribution to our knowledge of popular religion. But Schmitt includes Braudel’s longue durée, the long haul, as well, and here makes his most astonishing revelation, namely that the cult of St. Guinefort was still going strong in the nineteenth century. Schmitt found only one intervening trace, a document of 1632 referring to a chapel of St. Guy le Fort located right where he would place the original cult. But then, in 1826, a survey by the local bishop produced a letter about the spurious cult of St. Guinefort to whom mothers brought sickly children. The silence of six centuries was thus broken three times, once just to list the name of a chapel, and twice for extended comment, both times by clerics, who were as hostile to popular culture in the later period as in the earlier.


By the nineteenth century, however, the clerical monopoly on literacy had long since passed, and when the exemplum was published in 1877 an anti-clerical folklore expert, A. Vayssière, soon appeared on the scene. Everyone he spoke with avowed that Guinefort was a dog, and they told the story of how he died, with the noteworthy detail that the household was not a knight’s chateau but a woodsman’s cabin. A clerical reply in 1886, by the Abbé Delaigne, insisted on forcing the issue: “Is Guinefort a saint or a dog? That is the question.” Of course he proved to his own satisfaction that Guinefort was a saint, against the “free-thinker” Vayssière. This work was followed shortly by the (railroad) travel diary of the Baron Raverat, a firm believer in technological progress. He criticized the saint-or-dog debate as lacking any sense of historical evolution, and offered instead this neat formulation: “Guinefort was first a dog, then a saint.”

Finally, in the mid-1970s, Schmitt, like his urbane countryman of seven centuries before, went to the site of the cult of St. Guinefort to gather oral testimony. He recorded a dozen statements about an old woman still active in the 1930s who survived by begging, lighting candles for other people in church, and carrying out vicarious pilgrimages. She was thought to be the last guardian of the cult of Guinefort, the final successor to the marginal hag of Stephen’s exemplum.

Schmitt explains the continuation of this cult through six centuries by the stable economic and social conditions that prevailed throughout the same period. The unproductive lands that were imprudently given over to cultivation in the twelfth century and shortly thereafter withdrawn from it were then flooded to form fish ponds. These ponds stayed in place for six centuries and were not without effect on the health of the local inhabitants. In the 1840s their life expectancy was twenty-three years, as against a national average of thirty-five. Also in the 1840s, 90 percent of the young men called up for military service from that region were rejected, the leading reasons being “weak constitution” and small size. The draining and drying of the ponds began in 1836 and proceeded over the next half century, by which time the railroad arrived, bringing the Baron Raverat among others. The influence of the clergy gave way to that of learned lay people. The fish ponds dried up, and so did the cult of St. Guinefort.

Here indeed is the longue durée. What began as an incident in thirteenth-century ecclesiastical history is found to have roots in Indo-European culture and reverberations in the twentieth century. Those grand conventions of historiography, the Renaissance and the Reformation, have no place in this scheme of things, although is it surprising that no Counter-Reformation zealot smoked out the Guinefort cult.

In describing how Schmitt has brilliantly drawn on different disciplines to reconstruct the story of Guinefort, I have not even mentioned his explorations in iconography, archaeology, or astronomy. As a model for the study of the culture of preindustrial Europe his book could be as influential as Carlo Ginzburg’s The Cheese and the Worms6 and Emmanuel Le Roy Ladurie’s Montaillou.7

All three are based on flukes. Ginzburg discovered in the archives of the Inquisition the case of a miller from the Friuli who had developed a cosmology in which the universe was formed in the way that cheese is made. The miller was interrogated at length and finally put to death in 1559. Although the miller owned, borrowed, and apparently read a few books, Ginzburg argues that the most important source of the miller’s cosmology was oral; he drew on a cosmological tradition preserved through several centuries within peasant society. Both Ginzburg and Schmitt find that at the crucial point the oral tradition “from below” had great influence and they assert this conclusion perhaps a bit more categorically than some historians would find convincing. But both are brilliant detectives. If Schmitt’s book reads like a methodical police inspector’s report, Ginzburg’s, whether in Italian or in English translation, is a finely polished giallo.

Le Roy Ladurie exploited the scholarly edition of an Inquisition register about an out-of-the-way mountain village. He made of it as full a reconstruction of European village life as exists anywhere in modern historical writing. But to do this he had to slice the testimonies of individuals into numerous tiny pieces, which he then rearranged in structural categories suitable for studying village life. This led not only to frequent repetition but also to occasional transpositions that make the sources appear to say things that they did not and could not say. While Montaillou is rich in information on fourteenth-century pastoral society, it remains a unique curiosity. It tells us what was going on at 3900 feet above sea level and perhaps, by extension, it gives some idea of what went on in other, more important places before the Inquisition arrived. However, as a model, it is of limited value and in some of its particulars may even be misleading.

Schmitt’s discovery is founded upon a similar curiosity. Few literate people cared at any time what religious beliefs and practices prevailed in the barren countryside of the Dombes. But Stephen of Bourbon made of his investigation a parable for preachers. And Schmitt has investigated that parable in ways that hold much promise for other historians.

This Issue

April 30, 1981