The British Journal of Psychiatry
Suicide or accident - self-killing in medieval England
Series of 198 cases from the Eyre records


Background Little is known about suicide in England in the medieval period. Legal records provide the best source of post-mortem data about suicides.

Method Selected Eyre records from the reigns of Henry III (1216-1272), Edward I (1272-1307), Edward II (1307-1327) and Edward III (1327-1377) were translated and examined for details of self-killing.

Results One hundred and ninety-eight cases of self-killing were found, eight of which were found to be accidental, non-felonious deaths. Self-killing was more common in men. Hanging was the most common and drowning the second most common method of self-killing in both males and females. Self-killing with sharp objects was predominantly a male method. Other methods of self-killing were rare. There were no reports of deliberate self-poisoning. There is some evidence of underreporting of, and attempts to conceal, self-killing from royal officers.

Conclusions Eyre records suggest that although some of the facts surrounding self-killing have changed, others have remained constant, particularly the higher proportion of men who kill themselves and the greater use by men than women of sharp instruments to kill themselves. We discuss the description and understanding of psychiatric states by medieval English Eyres, particularly in terms of the perception of the mental states that accompanied suicidal actions.

The evolution of ideas about suicide in a legal and historical context should be relevant to psychiatry today. Examining historical records allows us to explore early attitudes towards suicide and the demographics of suicide in this period, adding to a longitudinal perspective of mental illness and self-killing.

Suicide in medieval England and Eyre records

Suicide in medieval England was both a religious and a secular concern. This is a particularly important time in the evolution of the common law and legal ideas about intent and responsibility, and is the earliest period from which central, official records on suicide are available. Self-murder was a mortal sin in the eyes of the Church, penalised by prohibition of burial in consecrated ground and also confiscation by royal authorities of the goods of the deceased and the implement used to commit suicide (the deodand), whether it belonged to the deceased or not.

The investigation of all deaths that did not seem to be natural was entrusted to the coroner and a jury drawn from men in the area in which the body was found. The jury was to report:

  1. who found the deceased;

  2. how the deceased died: whether of natural causes, by accident, by murder or self-murder, and whether the death was felonious (a serious offence resulting in the liability of execution of a living offender and to confiscation of land and goods after death), that is, committed when the deceased was responsible for his/her actions;

  3. whether the deceased had any chattels (goods other than land) and who had them now.

Reports of these investigations can be found in the few surviving rolls of coroners from the 13th and early 14th centuries and, in greater numbers, in the records of the Eyre, a judicial and administrative inquiry that was sent out to English counties from the 12th to the early 14th centuries. Drawing on information from coroners' inquests, Eyre juries were required to report all suicides that had occurred since the last Eyre (Hunnisett, 1959; Helmholz, 1986). Their information was not necessarily recent because Eyres visited counties no more than every 7 years (Barker, 1990). Reports on suicides contained in them are generally very terse, and are recorded in abbreviated, medieval legal Latin.

Eyre records are housed in the Public Record Office and are the best existing source of information about suicide in medieval England. There are few medical records for such an early period, the survival of coroners' rolls is extremely patchy and very few medieval English church court records survive (Hunnisett, 1959; Helmholz, 1986). A small number of references to suicide appear in chronicles; these are interesting but would be unusual cases and may not be representative of suicides in general in this period. Murray (1998) has examined and analysed some of the Eyre records for this period but does not entirely address issues of mental illness and self-killing.

Eyre records cannot answer all of the questions of interest to psychiatrists; in particular, they cannot be used to give reliable data on suicide rates. They can, however, be used to give useful qualitative data on suicidal behaviour and concepts of mental state and responsibility for one's actions.


A sample of surviving records of Eyres were examined, transcribed and translated by G.S. The majority of the records examined are from the period 1272-1300 because of the greater consistency of record-keeping in this period. Transcribed records were examined by A.S. for demographic data (gender, women's marital status, men's occupation), verdict as to the cause of death, site and method of suicide and wealth of the deceased (as shown by the value of the goods that were forfeited to the king).


Cases of self-killing were found in 28 rolls from 1227 to 1330 (Table 1). A total of 198 cases of self-killing were recorded: 135 male and 63 female. Of these records 188 cases were regarded as suicide (i.e. self-murder) that was deemed felonious. One case was regarded as not being a self-killing because it was a death due to self-starvation, but we have included this because it illustrates that omissions were regarded differently to active self-harm. Another case was initially regarded as being a suicide but later, when more information was available, was re-graded as an accidental death.

View this table:
Table 1

The verdict of Eyre rolls from selected English Eyre records between 1227 and 1330 (by gender)

Only eight cases of self-killing (six male and two female) were recorded as accidents - non-felonious acts - rather than being the deliberate, willed act of the deceased for which he/she was legally responsible. Various terms were used to describe the state of illness of the individuals found to have killed themselves but not to have been legally responsible for their action (which was regarded as not having been caused by them because they were ill). Mental illness is suggested by the reports, which describe the deceased as having been demens (having lost his or her mind) or as having acted in frenesi or per frenesy (in a frenzied state: through or because of frenzy) or furore detentus (taken by madness). Understanding of the effects on the mind of physical illness can be seen in descriptions of people having been “suffering from an illness” or “suffering from a high fever” at the time when the act causing death was committed.

All the cases of accidental self-killing had descriptions of the individuals having some sort of altered mental state involving a frenzy or suffering from an illness. Abnormal mental states are mentioned in only six of the cases of felonious suicides where the individuals are described as being in a frenzy, et demens (without sense or mind) or lunitacus.


The following transcriptions of case vignettes from the Eyre rolls illustrate a number of cases where different verdicts were found in self-killings.

Verdicts of suicide

“William de Wedmore, vicar of Chryriton hanged himself in his own home in the same village. The verdict was suicide. And Walter de Wedmore and John his brother(s) buried the said William without view of the coroner and took his chattels, value 34s so they are to be arrested. Afterwards they came to court and the sheriff let them go.” (Eyre of Somerset 1280)

This is a typical suicide verdict, the individual being judged to have killed himself feloniously with deliberate intent. It illustrates attempts of relatives to conceal a suicide in order to bury an individual, perhaps in consecrated ground, and take his goods, which ought to have been forfeited to the king, for themselves.

“A certain Richard son of Sibill gratis drowned himself in a pit at Marshton St Lawrence. No one is suspected. Judgment felonious suicide. Chattels 6s 2d.” (Eyre of Northamptonshire 1329)

This is a typical terse entry of a felonious suicide with no record of the village concealing the death.

Verdicts of accident

“William la Emeyse of this vill, suffering from an acute fever which took away his senses, got up at night, entered the water of Kentford and drowned himself. The jury was asked if he did this feloniously and said no, he did it through his illness. The verdict was an accident.” (Eyre of Hereford 1292)

In this case the individual was found not to have feloniously committed suicide because the cause of his death was his illness, rather than his own specific intent. As a self-killing without intent it was regarded as an accident.

“Thomas, son of Henry Robekyn of Brandon, in a frenzy (habens frenesium) cut off his left foot with an axe and then his left hand, in the house of the said Henry, his father, in Brandon, and the following night he died from this. The first finder and four neighbours came and are not suspected. The jury, asked if they suspect anyone else of this death say no and say that the said Thomas straight after the deed came to his senses (reddit' ad sensum proprium) and had the last rites of the church before he died. So the deed was done through loss of sense and not feloniously. So the judgment is accident.” (Eyre of Norfolk 1286)

Here, an individual was found to have done himself fatal injuries while he was unable to form a specific intent because he was in a temporary state of frenzy. The first finder acts as a witness to the Eyre. The verdict of accidental death (infortunium) was bolstered by the fact that before his death he was able to repent of his sins and have the last rites.

Verdict initially suicide, later revised to accident

“Ricardus de Kirkeby killed himself with his sword. First judged to be suicide and his chattels were forfeit. Later it was sworn that he accidentally fell on his sword.” (Eyre of Westmoreland 1278-1279)

In the margin the initial verdict of suicide is crossed out and ‘ accident’ is written. This might suggest that there was not an initial reluctance to find a verdict of suicide.

Verdict initially accident, later changed to suicide

“Robert le Pestur accidentally drowned himself. Later (after wife died) found to have done so voluntarily so verdict changed to suicide.” (Eyre of Devon 1238)

Suicide not considered

“William de Boubegh burgled the home of John Martyn of Blaneford and stole a bushel and a half of wheat and ran into the village watch who took him and put him in custody in the stocks at Bryanston for 8 days. And this William for this 8 days there would neither eat nor drink. After the said 8 days, he died there. And the jury asked if the said William died through any constraint or hardship whilst in prison. They said no. And it was recorded in the Eyre roll that the matter was investigated before the coroner.” (Eyre of Dorset 1288)

There is no record of a verdict of suicide here and the death appears to have been recorded because it was a death in custody. This shows that a strict distinction was made between causing one's own death by deliberate act and doing so by deliberate omission.

Demographic information derived from Eyre records

Marital status

It is not always possible to tell the marital status of women (and it is never documented for men unless their wife was the finder of their body). A woman is sometimes described as being the wife of a man but it is often not clear if her husband is alive or not. In strict law, the chattels of a woman came under the control of her husband while they were married, so if a woman is recorded as having had chattels, this implies that she was not married or was a widow (Barker, 1990). However, we have only used marital status noted in the records rather than making inter-pretations based on the presence of chattels.

One woman is described as being unknown to the jury and for another 23, no note of spouse or relatives is made. Of the remaining cases, 11 are described as being the daughter of a man, probably implying that they are not married, 4 are said to be widows and 24 are said to be the wife of a man. Although suicide is more common now in the separated, divorced and widowed (Kreitman, 1988), there are no data about the relative proportion of married, unmarried and widowed women in medieval society so it is not possible to determine whether any particular group is overrepresented in our group of suicides.


Occupation is recorded in few men - probably a reflection of the irrelevancy of this to the court but possibly because of the low frequency of specific occupations - but five clergymen, two servants, one goldsmith, two millers and one baker can be identified. There do not appear to be any individuals of really high rank because they would certainly have been identified by title.

Wealth of suicides

Goods forfeited to the king are recorded in all but two cases (Table 2). Some individuals only forfeited the deodand (weapon of their self-killing), whereas others had significant resources (8 pounds and 6 shillings in the case of one woman and 31 pounds and 2 shillings in the case of one man) and would be regarded as ‘ wealthy’ (Dyer, 1989). Women rarely forfeited more than a pound.

View this table:
Table 2

Goods forfeited to the crown by individuals committing felonious suicide, from selected English Eyre records between 1227 and 1330 (by gender)

Method of suicide

It was usual to record the method of suicide in the Eyre records, both because of a need to show the cause of death and because the implement causing a death was forfeited to the king under the law of deodand. In 189 cases the method of suicide was recorded. We have grouped deaths by stabbing or cutting with knives, daggers, forks, swords or other sharp objects together (Table 3). Hanging is the most common method of self-killing in both men and women, drowning is the next most common in men and women and the use of sharp objects is an almost exclusively male method of self-killing. Other methods of self-killing are rare.

View this table:
Table 3

Method of self-killing from selected English Eyre records between 1227 and 1330 (by gender and verdict)

Location of death

In two-thirds of records the general location of the suicide is noted: whether the body was found in its own home, that of another, in water or in another outside location such as a field or wood (Table 4). Given our current understanding of medieval gender roles, with women being largely based inside the home (Hanawalt, 1998), we expected that men might be more likely to commit suicide outside the home and women inside the home. There is, however, no gender difference in the site of committing suicide.

View this table:
Table 4

Site of self-killing from selected English Eyre records between 1227 and 1330 (by gender)

Time of suicide

In a small number of cases of suicide (one woman and five men) it was noted that the suicide took place at night. The time of suicide was not otherwise noted.


Reliability of records

Although the Eyre records are incomplete, we have managed to find records of 198 cases of self-killing. That so many records exist is a result of the importance of legal documents and of the early centralisation of justice in medieval England. English legal documents afford the best available body of factual information on medieval suicide.

Murray (1998), in the first of his volumes on suicide in the Middle Ages, examines medieval legal records, including some Eyre rolls, for the period that we have studied. Although we have a number of cases in common, we present new cases that have not been reported elsewhere and we have explored the information differently.

We would anticipate that the records that we have examined underrepresent the true number of suicides in the communities that the reporting Eyres represent for a number of reasons. First, there are clearly missing records for this period. Sometimes it can be seen that suicide cases mentioned at the coroner's inquest do not appear on the Eyre roll. Second, it is likely that suicide was underreported. It is recognised that even today suicide rates are underestimated through imperfect detection and an assumption that a death that might have been suicide was an accident unless proved otherwise (Symonds, 1985). It has been suggested that there would have been pressure on medieval communities to underreport suicides to royal officials, because of the consequences for the family of the suicide, who could not bury him or her in consecrated ground according to the rites of the Church, and would be unable to keep the suicide's goods (Hanawalt, 1979; MacDonald & Murphy, 1990). The records show some cases of attempts to conceal suicides, with families or neighbours burying those who had killed themselves, or removing the goods of a suicide before they could be examined by royal authorities. Presumably some similar offences went undetected. More detailed examination of the Eyres shows no evidence that self-killing was underreported to a greater extent than other offences: it is accepted that there was a gap between law and practice in medieval England.

The cases may be drawn from an unrepresentatively affluent population. No reliable information is available as to the population size and wealth in England before the late 14th century, but on any estimate the number of people with little or no money or valuable goods that could be forfeited to the king must have been large. Few poor individuals, however, appear in our data. It is possible, but unlikely, that poorer individuals were less inclined to kill themselves. It would be more likely that those responsible for reporting suspicious deaths to the coroner did not report deaths of poorer individuals without social standing or goods to be forfeited.

Although clerics appear to be over-represented, many men were involved in church activities and the presence of a description as a cleric does not indicate their level of involvement in the church. What does appear to be important in their description as a cleric is that their profession is mentioned at all, whereas few others are. Perhaps this is because the sin of self-murder was particularly abhorrent in those who were preaching that it was a sin. It might be suggested that there would be less pressure for concealment of suicide of clerics who did not have descendants to inherit their goods, but because some lower level priests were able to marry there would have been the usual pressure to conceal their suicide.

Methods of self-killing

As the implement causing death was forfeited to the crown, relatively complete records of method of suicide were available. There are no recorded cases of self-poisoning as a method of suicide. Self-poisoning was a particularly unacceptable method of self-killing and there may have been some cultural bias against using this method to kill oneself (MacDonald & Murphy, 1990). It seems unlikely, however, given the accessibility of naturally occurring poisons and widespread knowledge of herbal remedies (Rawcliffe, 1997), that self-poisoning did not occur. Chronicle sources have shown knowledge of poisoning in this period (Paris, 1880). More likely explanations of the lack of reports of suicide by poisoning would seem to be either the inability of medieval inquests to detect the presence of poisons in the body or the lack of suspicion aroused by deaths without the presence of physical evidence external to the body.

Hanging was the most commonly recorded method of suicide in both men and women, accounting for over half the deaths in total. It would be likely that a greater proportion of deaths by hanging than other methods were reported to the Eyre because there was incontrovertible physical evidence of self-killing and objects to be forfeited to the crown (the rope and beam). It may also be that there actually were proportionately more deaths by hanging. In women, it could be that the reluctance to hang themselves, which is suggested by later figures, was overcome in the Middle Ages because of a lack of alternatives. They may have been less able or willing to poison themselves and, having a more home-based domestic role, may have been less able to drown themselves because this would have necessitated them leaving their usual ‘women's space’. Our data, however, do not give us any evidence to support this.

Death by drowning was also common. It is likely to have been underreported because if the body was not found the coroner did not have to become involved. In addition, deaths by drowning often may have been thought to be accidental and not usually reported as suspicious.

The use of knives, axes or swords is predominantly a male method of suicide. This may reflect cultural acceptability of male use of such implements or greater accessibility of these for men. The latter seems unlikely, because women would have ready access to potentially lethal household knives and axes. The greater use by men than women of sharp instruments as a method of suicide may, rather, be due to the cultural acceptability of violence in men that might generalise to violence in suicide in men. This shows an important consistency with the excess of violent methods of suicide seen in men now (Appleby et al, 1999).

Self-burning is rare, as is death by jumping, reflecting the lack of high buildings and that jumping from cliffs or bridges into water might be regarded as an accidental fall, the body could have disappeared or, if seen as deliberate, might have been recorded as a drowning case.

The case of self-starvation and deprivation of fluid was not found to be suicide, implying that death by omission was not regarded as self-killing. The distinction between acts and omissions remains present in modern law, for example in cases of hunger strikers (Leigh v. Gladstone, 1909; Secretary of State for Home Department ex parte Robb, 1995).

Comparison with later suicide statistics

There are twice as many recorded cases of suicide in men than women. This may be a true record of gender ratios but it is possible that the rates in women are under-represented because married women were unlikely to have independent goods to be forfeited to the crown. However, one of the more robust findings in suicide statistics across the ages in England is the higher recorded suicide rate in men compared with women. Contemporary statistics and studies looking at suicide in Victorian, Tudor and Stuart times have all shown similarly increased rates in men (Anderson, 1987; MacDonald & Murphy, 1990). The roles of men and women in society have changed little until recent times so it is not possible to say whether this gender difference has a biological or social origin.

Although the excess of male suicides appears to be constant, the methods of suicide appear to vary with time, which probably reflects accessible and culturally acceptable means of suicide. In medieval suicides, as mentioned previously, there are no cases of poisoning because this was a particularly abhorrent method, whereas today this is a common method of suicide in women in Britain. In our cases, hanging is the most common method of suicide and there are no gender differences in this, whereas the Victorian studies suggest that men were much more likely to hang themselves than women, whose characteristic method of suicide was drowning.

Descriptions of mental illness and understanding of intent

Clarke (1975) discusses descriptions and treatment of those with mental illness in this period. The language used in the Eyre rolls to describe suicides apparently due to mental illness is consistent with that which Clarke discusses and in our study the use of descriptions of the felonious and non-felonious suicides in those apparently suffering from mental illness is interesting. Returning a verdict of accident, rather than suicide, in cases of self-killing was rare. Accidental verdicts were found where mental disturbance was caused by a physical illness (e.g. “labouring under an acute fever”) or an inherent mental disturbance did not allow the individual to form the requisite intent. In this case series there were only eight such accidental verdicts. These individuals had to be sufficiently impaired as a result of illness to be unable to form the intent to kill themselves. The descriptions of these individuals imply that they were suffering from severe mental illnesses, probably psychotic illnesses, and that they were easily recognisable as behaving unusually to those around them. Of the felonious suicides, six were identified as clearly suffering from mental illness but apparently able to be responsible for their own actions and thus for killing themselves. That the remaining 182 individuals killing themselves were free of any mental illness would seem very unlikely, given current information about mental illness in those killing themselves (Barraclough et al, 1974). It would seem more likely that many of them were suffering from non-psychotic or quietly psychotic illnesses. This implies that there was a difference in the understanding of the ability to form intent, that labouring under a frenzy one could not do so whereas in other psychiatric states one could. Although not as yet systematically examined, homicide records from this period would seem to support this distinction (Hunard, 1969).

Significantly, in the eight cases of non-felonious self-killing there are no cases of hanging, although it is common in the sample as a whole. This may shed more light on the idea of the sort of mental states that were regarded as excusing conduct that might otherwise be criminal. The act of hanging may have been seen as requiring a more deliberate plan than self-killing by drowning or stabbing, which might be more impulsive and require less structured thought, so that it was inconceivable to medieval jurors that a person whose mental state was such that it would excuse otherwise criminal actions would be able to hang himself.

Further research in this field will focus on the attitudes of juries during this period, an extension of the examination of coroners' records into the late medieval period, many of which have not yet been examined, and exploring the evolution of ideas about intent and psychiatric theories. In addition we intend to develop an understanding of across-time, cross-cultural psychiatry. There has recently been an upsurge of interest among historians and lawyers in suicide, and this area should also be of importance to and illuminated by psychiatrists.

Clinical Implications and Limitations


  • This paper gives a historical perspective on suicide at a time of evolving legal and religious attitudes to suicide during the earliest period for which English legal records are available.

  • Self-killing is more common in men than women in the medieval period; this is a consistent finding over time.

  • Few of the individuals killing themselves were regarded as suffering from mental illnesses.


  • Some Eyre records are clearly missing from this period.

  • Self-killings are likely to have been underreported to the Eyres.

  • Recording of self-killings is brief and incomplete.

  • Received June 11, 1999.
  • Revision received June 26, 2000.
  • Accepted June 28, 2000.


View Abstract