Top of page

Swimming a Witch: Evidence in 17th-century English Witchcraft Trials

Share this post:

In recent posts on this blog, I have written about the evidence used in 17th-century witch trials, both in America and in England. In those posts, I pointed out that proving the crime of witchcraft was no simple matter. Rules for evidence in criminal trials were not yet formalized at that time, and opinions about what constituted proof of things supernatural varied widely among judges, writers, and the wider population. This post is about a strategy to prove an accusation of witchcraft that came into use in England in the early Modern Era: “ducking” or “swimming” a witch.

Witch swimming was the practice of tying up and dunking the accused into a body of water to determine whether they sink or float. Sinking to the bottom indicated that the accused was innocent while floating indicated a guilty verdict. (Kittredge, p. 232.)

The practice had its origins in Western Europe as early as the ninth century and spread with the rise of Christianity throughout the continent. (Bartlett p. 10-12.) In the Middle Ages, it was known as the ordeal of the cold water, or judicium aquae frigidae. (Kittredge, p. 232.) It appeared in the laws of England in the Laws of Aethelstan (ca. 930 AD) (Lieberbmann, I, p. 162); and of Aethelred (ca. 1000) (Liebermann, I, p. 230); and of William the Conqueror (ca. 1100) (Liebermann, I, p. 502-503). (Kittredge, p. 233) Over twenty instances are recorded in the Pipe Rolls in the year of the Assize of Clarendon (1166), and about that many in the year of the Assize of Northampton (1176) among other references. (Kittredge, p. 234.)

The title page of John Cottas pamphlet The Trial of Witchcraft London, 1616.
The title page of The Trial of Witchcraft London (1616) by the Northampton physician John Cotta. Cotta believed the swimming test sought a miracle where there was none. Nature, he argued, did not distinguish between good persons and bad ones. Photo by Nathan Dorn.

The premise of the ordeal was that the trial provoked God’s direct intervention in determining the guilt or innocence of the accused. Since the ordeal’s result was a revelation of God’s judgment, the ordeal was often called judicium dei, the judgment of God. (Bartlett, pp. 162-166.) People seem to have taken this aspect of the custom seriously. Administration of the ordeal was undertaken by clergy. Its execution included liturgical formulas, the blessing of the elements of the ordeal, and the celebration of Mass. (von Schwerin, pp. 42-48.)

While this custom may itself seem like a magical act – and some later critics, especially in the 17th century held it to be so (Darr, p. 164) – the medieval ordeal was answering a need. There were some crimes that were simply not amenable to other forms of judicial proof; namely, crimes for which there was no witness, such as treason, adultery, witchcraft, or heresy; or those for which the defendant could not find compurgators, or for which the defendant’s oath was held to be unreliable. (Bartlett, pp. 30-33.) In these cases, what was needed was evidence from an absolute, or divine, point of view. The ordeal purported to supply that point of view in a way that many, though certainly not all, people of the time and place found consonant with their beliefs. (Bartlett, pp. 162-166.)

As for its opponents, the most important arguments raised against the ordeal were, first, that they were uncanonical, lacking any secure foundation in church law or biblical texts; and second, that they were an unwarranted tempting of God. Both of these arguments were made by critics of the custom as early as the ninth century. But it was not until the 13th century that they became the prevailing opinion of the papal party. It was decided in the Fourth Lateran Council (1215) that clergy were to withdraw their support from the ordeal. (McAuley, p. 474.) This led to widespread abandonment of the custom in the following decades; England’s King Henry III declared ordeals illegal in 1219 in a royal mandate to the justices. (Patent Rolls, 1, p. 186.)

It is not altogether clear whether the custom survived or was resurrected in early Modern Europe. Instances of the swimming test start to appear during the witch hunts that took place in late 16th-century Europe and Scotland. (Bartlett, p. 146.) King James VI of Scotland was much affected by these events and was personally horrified by revelations that an apparent conspiracy of witches sought to murder him and his family during a sea voyage by means of a collective magical working; afterward, he wrote a treatise on the reality and menace of witchcraft Dæmonologie, in forme of a dialogue, divided into three Bookes, which he published in 1597. The book was published again in 1603 after James became King of England. In it, he pointed out that the swimming test is a “good help” in confirming witchcraft accusations. He proposed that the reason it works is that since witches have by their dealings with the devil refused the benefit of baptism, that water, which is the element of baptism, will effectively spit them out and prevent them from sinking: “it appears that God has appointed (for a supernatural sign of the monstrous impiety of the witches) that the water shall refuse to receive them in her bosom, that have shaken off them the sacred water of Baptism, and willfully refused the benefit thereof.” (James I, p. 80.)

Title page of Richard Bernard, A Guide to Grand Jury Men London, 1627)
Richard Bernard argued in his pamphlet A Guide to Grand Jury Men (London, 1627) that the swimming test amounted to an impious attempt to demand a miracle from God. Photo by Nathan Dorn.

The same explanation of how the swimming test works appears in a pamphlet that records the first known instance of the test’s use in 17th-century England, The Witches of Northamptonshire (London, 1612). (Darr, p. 161.) The assize judges that applied it in that case were likely encouraged by King James’ Dæmonologie as well as the 1603 Witchcraft Act (which criminalized witchcraft but did not relate to the swimming test) that James I enacted in his first year as King of England.

The swimming test was used frequently throughout the next several decades but it was never embraced by learned authorities. (Darr, p. 158.) The Puritan theologian William Perkins, in a pamphlet published in 1608, rebutted the notion that water rejected the witch’s body, saying that not all water is the water of the baptism. The swimming test and similar customs were, “after a sort, practices of witchcraft, having no power by God’s ordinance.” (Bartlett, p. 148.) The physician John Cotta argued in a 1616 essay about the scientific and theological aspects of witchcraft that people who used the test failed to grasp that the natural world follows immutable laws. Nature, he explained, does not distinguish between good persons and bad ones. In 1627, the Puritan clergyman Richard Bernard wrote about the test in a short but influential book which he called A guide to grand-iury men: diuided into two bookes: in the first, is the authors best aduice to them what to doe, before they bring in a billa vera in cases of witchcraft, (London, 1627). In that book, he advised that the it was illegal, and, echoing medieval critics, that its use amounted to an impious attempt to demand a miracle from God. (Darr, p. 164.)

Generally, justices of the assizes appear to have opposed the test. It was more likely to be applied in county courts by justices of the peace, whose ranks were composed of untrained local gentry. (Sharpe, pp. 218) Between March and May 1645, Matthew Hopkins (the so-called Witchfinder General) and John Stearnes regularly used the test until assize judges arrived and ordered them to stop. (Sharpe, p. 218.) Many authorities in Europe were also critical of its use. (Bartlett, p. 147.) By the end of the 17th century, courts were apt to consider witch swimming as an assault committed against its target; and if the subject were to drown during the procedure, it was a murder. The custom was always connected to popular witch lore, and its execution was often a thinly concealed act of mob justice. As time went on, it was more often simply a lynching; a striking example of this was the case of Ruth Osborne, who died in 1751 – 16 years after the Witchcraft Act of 1735 (9 Geo. 2 c. 5) made it a crime to make accusations of witchcraft –  when a man named Thomas Colley violently drowned her while performing the test in front of a crowd of 5,000 onlookers. Colley was hanged for her murder at the Hertford assizes on July 30, 1751. (Sharpe, pp. 1-4.)

For more In Custodia Legis posts on witchcraft look here.

Sources in the Library of Congress related to this subject:

Ady, Thomas. A candle in the dark; or, A treatise concerning the nature of witches & witchcraft; being advice to judges, sheriffes, justices of the peace, and grand-jury men, what to do, before they passe sentence on such as are arraigned for their lives, as witches. London, Printed for R. I. to be sold by T. Newberry, 1656.

Bernard, Richard. A guide to grand-iury men: diuided into two bookes: in the first, is the authors best aduice to them what to doe, before they bring in a billa vera in cases of witchcraft… London: Printed by Felix Kingston for Ed. Blackmore …, 1627.

Cotta, John. The triall of witch-craft: shewing the true and right methode of the discouery, with a confutation of erroneous wayes. Printed by George Purslowe for Samuel Rand and are to be solde at his shop …, 1616.

Filmer, Robert. An advertisement to the jury-men of England touching witches. Exeter: The Rota, 1975.

Perkins, William, 1558-1602. A discourse of the damned art of witchcraft, so farre forth as it is reuealed in the Scriptures and manifest by true experience. [Cambridge, Cambridgeshire]: Printed by Cantrell Legge …, 1618.

Scot, Reginald. The discouerie of witchcraft… [London] By [Henry Denham for] William Brome, 1584.

Summers, Montague. The discovery of witches; a study of Master Matthew Hopkins, commonly call’d Witch finder generall, by the Rev. Montague Summers, together with a reprint of The discovery of witches from the rare original of 1647. London, The Cayme Press. 1928.

Secondary Sources:

Baldwin, J. “The Intellectual Preparation for the Canon of 1215 against Ordeals,” (1961) 36 Speculum 613-36.

Bartlett, Robert. Trial by fire and water: the medieval judicial ordeal. Oxford: Clarendon Press, 1986.

Cockburn, J. S. A history of English assizes, 1558-1714. Cambridge [Eng.] University Press, 1972.

Darr, Orna Alyagon. Marks of an absolute witch: evidentiary dilemmas in early modern England. Farnham, Surrey; Burlington, Vt.: Ashgate Pub., c2011.

Gaskill, Malcolm. Witchfinders: a seventeenth-century English tragedy. Cambridge, Mass. : Harvard University Press, 2005.

Gaskill, Malcolm. “Witchcraft and Evidence in Early Modern England.” Past & Present, No. 198 (Feb. 2008), pp. 33-70.

Grässe, Johann Georg Theodor, 1814-1885. Bibliotheca magica et pneumatica, oder, Wissenschaftlich geordnete Bibliographie der wichtigsten in das Gebiet des Zauber-, Wunder-, Geister- und sonstigen Aberglaubens vorzüglich älterer Zeit einschlagenden Werke. Leipzig : Verlag von Wilhelm Engelmann, 1843.

Heikki Pihlajamaki, “Swimming the Witch, Pricking for the Devil’s Mark: Ordeals in the Early Modern Witchcraft Trials,” Journal of Legal History, 21, no. 2 (August 2000) pp. 35-58.

Langbein, John H. “Historical Foundations of Evidence Law” Columbia Law Review, Vol. 96, No. 5, (June 1996)

pp. 1168-1202.

Liebermann, F. (Felix). Die Gesetze der Angelsachsen / herausgegeben im Auftrage der Savigny-Stiftung von F. Liebermann. Halle: Max Niemeyer, 1903-16.

Sharpe, J. A. Instruments of darkness: witchcraft in early modern England. Philadelphia: University of Pennsylvania Press, c1996 (1997 printing).

Witches and witch-hunters. A reprint of A treatise of witchcraft, by Alexander Roberts (1616); The discovery of witches, by Mathew Hopkins (1647); [and] Scottish witchcraft trials, by J. W. Brodie (1891). With a foreword by A. E. Green. [Norwood, Pa.], Norword Editions, 1973.

Von Schwerin, Claudius Freiherr. “Rituale für Gottesurteile” Heidelberger Akademie der Wissenschaften. Philosophisch-historische Klasse. Sitzungsberichte (1932/1933) 3 Abhandlung.

Subscribe to In Custodia Legis – it’s free! – to receive interesting posts drawn from the Law Library of Congress’s vast collections and our staff’s expertise in U.S., foreign, and international law.


  1. Wasn’t there a specific writ to initiate witchcraft prosecutions?

Add a Comment

Your email address will not be published. Required fields are marked *