Working With Sources On Witchcraft In Early Modern Europe
The biggest mistake I see people make when studying Witchcraft In Early Modern Europe is starting with secondary literature. It doesn't help that many online databases organize trial records by modern taxonomy rather than by actual document structure. You need to go directly to the trial transcripts, court ledgers, and examination records whenever possible. The published compilations often smooth over the really interesting bits. I spent roughly three years pulling together a regional analysis of witch trials in the Holy Roman Empire. The core problem was that most trial records exist in scattered local archives across Germany, Switzerland, and Austria. There's no single index that covers everything. My workaround was to use the Witch Trials in Early Modern Europe Database at the University of North Texas as a starting point, then cross-reference every citation against the original archive holdings. That's how I found discrepancies. A trial listed as concluded in 1629 in one secondary source had actually dragged into 1631 in the court ledger at the relevant Stadtarchiv. Those extra two years changed the entire demographic picture for that region because the persecution wave hit later than the literature suggested.
Researching Witchcraft In Early Modern Europe
Primary sources break into a few categories, and each requires a different reading strategy. Trial transcripts are the main material, but they come in forms that vary wildly by jurisdiction. Some regions produced carefully written court records. Others have only the barest summaries. The confessions are where things get tricky because torture was commonly used, which means you're reading what someone said under duress, not necessarily what happened. I've learned to read confessions backwards — start with the final written version and work toward the earliest interrogation notes. The details usually get more specific the earlier they appear, before the interrogators started steering. Accusation lists and denunciation records are another category many researchers overlook. These are the initial complaints that triggered investigations. They're shorter and often contain raw information that got cleaned up during the trial process. Marriage records, property transfers, and tax documents sometimes show up as context too. A neighbor dispute over a cow can escalate into a murder accusation by winter. I found that keeping a spreadsheet tracking the relationship between accuser and accused before the trial started revealed patterns that never showed up in the trial documents themselves. Most feuds originated from property boundaries, inheritance disputes, or refused favors. The magic accusation came later as a way to frame harm that otherwise had no legal recourse. Medical and autopsy records occasionally survive, particularly in places like Scotland where post-mortem examinations sometimes included notes on supposed witch marks. These are rare but important because they show how physicians and legal authorities negotiated between empirical observation and supernatural explanation. The standard practice was to look for a "witch mark" — a sensitive spot on the body that didn't bleed or feel pain when pricked. I once spent a morning trying to verify whether a particular mark described in a 1590 Scottish trial was anatomically plausible. It turned out the examiner had described a mole near the shoulder, which is entirely normal, but the legal culture of the time treated any unusual skin marking as evidence. That's the interpretive gap you have to navigate constantly.
Printed pamphlets and broadsides from the period are accessible but require careful handling. They're sensationalized by design. Publishers wanted sales. Still, they're valuable for understanding what the general public thought about witchcraft at the time, separate from what courts or theologians believed. Comparing a broadside account with the corresponding trial transcript usually shows massive differences. The broadside often adds demonic details that never appeared in any court record. For actual access, the Witch Trials in Early Modern Europe Database at the University of North Texas is the most comprehensive starting point. It's free and searchable by date, location, and outcome. The Regensburg Witch Trials Database provides detailed records for one of the most intensively studied cities. For broader coverage, the European Witches' Trials Database projects across various universities are worth searching separately because their coverage areas don't always overlap. Many local German archives have digitized portions of their holdings now, but the finding aids are often only in German and poorly indexed. I've found that learning basic archival German and knowing how to navigate Inventare Online saves weeks of dead ends. The common pitfall is assuming that conviction rates tell you much about how common witchcraft actually was. Most historians estimate that actual magical practice, if it existed at all in the way contemporary people understood it, was exceedingly rare. What was common was conflict. The witchcraft accusation was a weapon in ordinary social warfare. The peak persecution periods in the late sixteenth and early seventeenth centuries correlate strongly with economic stress, war displacement, and religious polarization, not with any increase in actual harmful behavior. The number of people executed for witchcraft across all of Europe during the early modern period is estimated between 40,000 and 60,000. Most were women. About twenty to twenty-five percent of all accusations resulted in execution, which sounds low until you consider that many accused people died in prison before trial, and that some regions had execution rates above sixty percent during active persecution waves.
Get the Full Details

One counter-intuitive finding that emerged from the regional databases is that communities with stronger municipal courts and more bureaucratic oversight actually had fewer convictions. Where justice was administered by visiting judges or ecclesiastical courts with less local accountability, persecution moved faster and killed more people. The more institutionalized the legal process, the more likely it was to produce delays that allowed accused people to survive. This isn't obvious from the surface of the sources. It only shows up when you map conviction rates against institutional type across hundreds of cases. Another thing that beginners miss is the regional variation in legal standards. Some jurisdictions required two eyewitness accounts of actual magical harm for a conviction. Others accepted spectral evidence, which meant a accusation based on what the accuser claimed to have seen in a dream or vision. The acceptance of spectral evidence varied enormously by time and place. Scotland was more skeptical of it than continental Europe. Certain German territories banned it outright in the mid-seventeenth century after realizing it produced too many false convictions. The shift away from spectral evidence is one of the clearest signs that legal authorities themselves were losing confidence in witchcraft prosecutions, even before the intellectual debates of the Enlightenment became relevant. When you're doing this work, keep in mind that the sources are uneven. Some communities have dozens of trial records. Others have none at all, not because no trials happened, but because the archives were destroyed in wars, fires, or reorganization. The absence of evidence is not evidence of absence. I've seen researchers claim a region had no witchcraft persecution simply because the local archive was burned in the Thirty Years' War. That's not a conclusion you can draw from missing records.
The practical side of reading these sources involves getting comfortable with early modern handwriting, which varies significantly by region and decade. Kurrent and Sütterlin scripts appear in German sources. Latin abbreviation conventions are standard in ecclesiastical records. If you're working primarily in English sources, you'll deal with Secretary Hand, which has its own particular difficulties. Learning to read the script takes time. It's easier than it looks once you've seen enough examples, but it's the single biggest barrier for people who jump in without preparation. There's also the question of terminology. The words used in the sources don't map neatly onto modern concepts. Zauberei, hexerei, maleficium, veneficium — each carries different legal and theological connotations depending on when and where it was used. A text calling something maleficium is making a specific legal claim about harmful magic. Calling it divinatio is a different accusation with different legal consequences. Treating all of these as interchangeable is a mistake that shows up frequently in introductory surveys. If you're looking to go deeper, the Witchcraft Resources Bibliography maintained by historians in the field is one of the most useful reference tools available. It's continuously updated and covers journal articles, books, and some primary source editions. The Sixteenth-Century Journal and Early Modern History Sourcebook projects also have useful collections of translated documents. Reading the documents yourself, even in translation, is where you'll find the substance. Secondary sources are useful for orientation, but they tend to generalize in ways that obscure the actual complexity of what happened.