It is easier to discuss the age of women prosecuted for witchcraft in 17th century continental Europe (see previous blog post) than in England because their age was sometimes specified in court records. This was very rarely the case in English records. Women’s marital status was, however, generally given, so ‘wife’ and ‘widow’ can be used as rough proxies for ‘younger mature woman’ and ‘older woman’ respectively (though young widows were much more common than today, because of the higher mortality at all ages).
The easiest figures to interpret come from the Northern Circuit of the main criminal courts (Assizes) in the second half of the 17th century, and these suggest that between twice and three times more wives were accused of witchcraft than widows:
Out of 48 suspects, there were 23 wives, 8 widows, 1 girl, 2 men, and 14 women whose marital status was not specified.
Even if the latter were all widows, this would only equalise the proportion of wives and widows.
48 suspects are not very many, however. The most complete series of criminal records – indictments for the Home Circuit of the Assize courts – covers the entire period (1563-1736) when witchcraft was a capital crime in England. Despite their abundance, however, these records are not easy to interpret. They are bedevilled by large numbers of ‘spinsters’ – no fewer than 45% of 423 women prosecuted for witchcraft (see table below)! These cannot ALL have been single women (as we might assume today) because the proportions found elsewhere in early modern Europe and colonial America were nowhere near as high. These include 16% of women prosecuted for witchcraft (of 909) from various places in Europe collected by Brian Levack, 2% (of 319) in Scotland, and 8% of women and men (of 114) in New England. In their studies of witchcraft in Germany, Roper and Rowlands note that most of the women accused of witchcraft were, or had been, married (Rowlands), or had borne children (Roper).
The likelihood that most of these English women were unmarried becomes even more remote if we consider the additional 18% of women who were described on their indictments as ‘wife and spinster’ (see table). The term ‘spinster’ in fact referred to a female spinner (just as ‘brewster’ was a female brewer), and probably did not carry implications about marital status before about the middle of the 17th century. The occupational meaning is spelt out in a mid-16th century complaint by a woman (accused of murder) that she had been given the addition of wife and spinster on the indictment when she was a gentlewoman, and so should have been referred to as such. Gentry were identified by social status not occupation. Sir Edward Coke later declared (around 1630) that a gentlewoman could cause an indictment to be thrown out if it referred to her as a ‘spinster’. Marital status was clearly irrelevant to this issue.
This does not take us very far, however, in trying to explain why so many accused women were referred to as ‘spinsters’. Checking out a legal technicality takes us part of the way.
The criminal indictment, written on a narrow strip of parchment, followed a strict legal formula, which included the status or occupation of the accused. If neither of these were specified, the indictment could be rejected on technical grounds (resulting in the accused being reprieved). Clerks would even invent occupations, on occasion, to avoid this eventuality. Married women were usually simply referred to as ‘wife of …’, but the publication in 1581 of an earlier case (1525) raised some doubts as to whether an occupation was also required for them. Most women could not claim to have a specific occupation – apart from the multitude of tasks carried out by a housewife – but all women from the non-gentry did spinning, for their own household if not for payment. The thread for all bed and table linen, household cloths and aprons was spun in someone’s household, then woven into cloth by a local weaver. So it was not entirely fictitious to refer to all wives from the non-gentry as ’spinsters’, and this simplified the clerks’ work. Wives were never given any occupation other than ‘spinster’, even though they were often involved in a variety of money-making activities.
In the event, it was later confirmed in the court of Queen’s Bench (in 1590 and 1600) that an occupation was not required for wives, but nervous clerks often included the term ‘wife of AB, spinster’ just to be on the safe side.
The huge proportion of ‘spinsters’ prosecuted for witchcraft in the 1580s and 90s (82% of women) is presumably related to the legal issue raised in 1581. If so, it looks as if wives were simply being referred to as ‘spinsters’, because the number of references to ‘wife and spinster’ actually goes down:
Some of these ‘spinsters’ may have been unmarried women – law dictionaries from the middle of the 17th century reported that they were being referred to by this term. The term was no doubt convenient for the increasing number of never-married mature women for whom no other English word was available – ‘maid’ had been used for a single woman, but this also implied that she was young. But there was no assumption that the term applied exclusively to unmarried women, as it is defined today.
Were widows also referred to as ‘spinsters’? It would seem to make sense, except for two considerations. The first is that the term ‘widow and spinster’ was never used on any of the indictments considered here, implying that this extra term was not required for them. The second concerns the possible reason why courts were so concerned about whether or not accused wives should be given an occupation – they may have wanted to deter them from claiming that their husbands were implicated in the crime.
In principle, wives could be exonerated if they carried out a crime with their husbands, because it was their wifely duty (as femmes couvertes) to do as their husbands told them. In practice also, they were sometimes reprieved – at least if the crime was not murder or treason on the one hand, or some petty misdemeanour on the other. But the situation was not clearcut – and a woman often claimed ‘marital coercion’ even if it was clear that her husband knew nothing about her criminal activity. So it could perhaps be argued that giving her an occupation set her apart from her husband, giving her a more independent identity, and might deter her from trying to implicate him.
This issue would not have arisen in the case of widows, who were considered to be independent agents. So nothing would have been gained by giving them an occupation. On balance, therefore, those referred to as ‘spinsters’ were more likely to be wives or single women than widows – although the situation is admittedly not clearcut.
These considerations seriously complicate our task of comparing the proportions of wives and widows prosecuted for witchcraft in England. One approach to the problem is to exclude the figure for the 1580s-90s from our calculation, when there was such a very large proportion of ‘spinsters’. This is also unfortunately the period when witchcraft prosecutions were at their height, but we are still left with 2/3 of the total. The figures for the remaining decades are given in the table. It can be seen that a third of those prosecuted were wives, with widows and ‘spinsters’ accounting for just over a quarter each:
No. % Wife 109 34 Widow 92 28 Spinster 82 25 Man 37 11 Daughter 1 0 blank 4 1 Total 325 99
You may be wondering why the ‘spinster’ issue did not arise in the discussion of the records of the Northern Circuit. This is because the information was not taken from the indictments, which had to follow the strict legal formula discussed above, but from the evidence of witnesses and examinations by magistrates. The evidence and examinations have not survived for the Home Circuit, and the indictments for witchcraft on the Northern Circuit are difficult to disentangle from all the other strips of parchment bundled into boxes in the National Archives. They await their cataloguer, performing another great service for historians like that provided by Ewen for the Home Circuit nearly a hundred years ago.
So all in all, I think one can probably conclude that wives were at least as likely to be prosecuted for witchcraft as widows in the central criminal courts of early modern England. This suggests that the popular stereotype of the witch as old hag would have seriously misrepresented at least half of the accused, just as it did for the European witches discussed in my previous blog post. Perhaps, like them, the accused English women were moving just beyond their prime, losing influence in the community, when those who felt they had suffered at their hands moved in to strike.
 Covering Cumberland, Durham, Lancashire (though in practice there are no records for this county), Northumberland, Westmorland and Yorkshire. Many of the depositions (witness evidence) and examinations of suspects (but not the interrogatories – i.e. questions put to the latter) are printed in J Raine Depositions from the Castle of York, relating to offences committed in the northern counties in the 17th century. Surtees Society vol. 40 (1861). Originals in National Archives ASSI 45: 1/5/38-9; 3/1/242-4; 3/2/81; 3/2/129-5; 4/1/109-11 and 131; 4/2/12-14; 5/1/30-9 and 87-8; 5/2/30-31; 5/3/10-14 and 132; 5/5/1-4; 5/7/95; 6/1/69, 88-90 and 164e-168, and 134-5; 7/1/6-7, 59-61, 107 and 9-10, and 185-8; 7/2/62 and 103; 8/2/34; 9/1/144; 9/3/124; 10/2/80-84; 10/3/34-54 and 124-6; 11/1/90-93; 12/2/6-7; 12/4/55; 14/1/151; 16/3/54-56.
 Covering Essex, Hertfordshire, Kent, Surrey and Sussex. Printed in Ewen, C L’Estrange Witch Hunting and Witch Trials: the indictments for witchcraft from the records of 1373 Assizes held for the Home Circuit AD 1559-1736 (1929).
 BP Levack. The Witch-Hunt in Early Modern Europe (2006) , 3rd edition, p.155 (the places are: the City of Toul, Basel, Montbeliard, Geneva, Venice and Sweden); JP Demos Entertaining Satan: witchcraft and the culture of early New England (1982), p.72 (he excludes the Salem trials); L Roper Witch Craze: terror and fantasy in baroque Germany (2004), p.161; A Rowlands Witchcraft Narratives in Germany: Rothenburg 1561-1652 (2003), pp.170-71.
 JH Baker ‘Male and married spinsters’ Am. J. Legal Hist. 21 (1977), p.257; Coke published the decision in the Second Part of his Institutes of the Laws of England, which he was working on around 1630. ODNB.
 Carol Wiener put forward this argument in 1976, but assumed that ‘spinster’ meant ‘single woman’, so the implications were rather different. CZ Wiener ‘Is a spinster an unmarried woman?’ Am. J. Legal Hist. 20 (1976), 27-31; see also VC Edwards ‘The case of the married spinster: an alternative explanation’ Am. J. Legal Hist. 21 (1977), 260-5.
 G Walker Crime, Gender and Social Order in Early-Modern England (2003), pp.201-5; see also Edwards op.cit. pp.262-4. A similar protection for the husband had been available in some towns when his wife traded independently of him. Her status could be redefined as, in effect, a single woman – femme sole – and if she incurred debts they would not be taken from his goods (even though all her property was otherwise her husband’s). She could be imprisoned if unable to pay. In some circumstances this status benefited wives, but given that they more often claimed that they did not have femme sole status than the other way round, the benefits seem to have more often accrued to their husbands. So in this case as well as that of crimes, a legal fudge protected the husband from misbehaviour on the part of his wife. See MJ McIntosh, ‘The benefits and drawbacks of femme sole status in England, 1300-1630’ J. British Studies 44 (2005), 410-38.