Why they stole: women in the Old Bailey, 1779-1789.

MacKay, Lynn

Why did women steal? In his answer to this question for early-modern England, John Beattie noted that women turned to theft "for the same reason men stole in this period - largely as a means of survival, as a way of supplementing inadequate wages or of supplying the most basic wants."(1) While need was undoubtedly a major factor in most female theft, the motivations of these women were actually quite complex, and in some ways differed significantly from the reasons men had for stealing. As Garthine Walker has recently pointed out, "the dynamics of female criminality ... reveal a far more complex and instructive view of gendered experience than historians of crime have hitherto acknowledged."(2) The differences between male and female theft surfaced with respect to what they stole, but were even more apparent when the social contexts in which plebeian women and men operated are considered. These contexts, in turn, had a crucial impact on the motivations each sex had for stealing.

The complexity of women's motivations can be seen through an examination of the Old Bailey Sessions Papers from the 1780s. This was a seminal decade in the development of modern attitudes toward crime and punishment. There was a huge and sustained increase in the number of individuals charged with theft during these years (see Table One), and not surprisingly, the decade has figured prominently in historical studies of changing crime rates. Many upper-class contemporaries were deeply concerned, and a number of social commentators began to call for an overhaul, particularly to the system of punishment. Crime - especially theft - was thought to be a very serious social problem.(3)

There were also major difficulties in this period: 1780 to 1782 were, of course, war years. When the American War ended in 1783 some 130,000 soldiers and sailors were very rapidly demobilised. As well, 1782 and '83 saw poor harvests, and in the latter year an economic slump and high food prices made life difficult for many in the labouring classes. Toward the end of the decade there was another period of economic difficulty which peaked in the crisis of 1788.(4) This was made worse by severe weather, particularly in the winter of 1788-89 when a frost lasted from November until January and the Thames froze over.(5) According to L.D. Schwarz, real wages in London fell by just over 11 per cent between 1780 and 1790.(6) This was, without doubt, a difficult decade for many on the bottom rungs of the income ladder.

The Old Bailey Sessions Papers (hereafter OBSP) constitute one of the few verbatim accounts we have of eighteenth-century trials.(7) The figures in the analysis that follows are based on the legal year which began in December.(8) The OBSP are not complete trial transcripts; discussions of legal points were often truncated or omitted altogether. As the decade wore on and the number of indictments rose, there are more and more of what John Langbein has called 'squibs.'(9) These very truncated accounts, usually of more minor cases, could consist of little more than the charge, the verdict and the sentence. Nevertheless, there are in each year a more than sufficient number of cases - never fewer than 190 males and 89 females - containing the statements of prosecutors, witnesses and defendants to make viable the kind of analysis I propose to do.(10)

Table 1

THEFTS IN THE 1780S

YEAR MALES FEMALES TOTAL

1779-80(*) 227 214 441
1780-81 324 195 519
1781-82 355 214 569
1782-83 525 230 755
1783-84 738 211 949
1784-85 752 202 954
1785-86 699 216 915
1786-87 648 227 875
1787-88 538 175 713
1788-89 631 174 805

 5,437 2,058 7,495

* These dates refer to the legal year which began in December.

Various theft charges have been included in this analysis: stealing, stealing from a dwelling house, from bleaching grounds, from ships, stealing privily (which includes picking pockets and shoplifting), horse theft, sheep stealing, as well as burglary and housebreaking. Since robbery was almost always accompanied by assault charges, these offences have not been considered in the tabulations that follow, although at certain points robbery will be discussed. Finally, counterfeiting charges have also been excluded since these offences often seem to have been committed by professional gangs.(11)

The characteristics of women's theft were both similar to and different from male patterns. As Garthine Walker has shown for an earlier period, the assumptions that women were for the most part the helpmates of male criminals and that they stole goods of lesser value than did men, simply do not hold.(12) In the first place, the propensity to form criminal associations was virtually identical for both sexes. If the decade is treated as a whole, 28.7 per cent of both male and female thieves acted with accomplices. Obviously women were not any more likely to act in consort with others than were their male counterparts. Indeed, if robberies are added to the other kinds of theft under consideration, then males were more likely to act in collusion with others. Including robbery, 37.1 per cent of males and 32.3 per cent of females acted in association with others.(13) Nor were women likely to take items of lesser value than were male thieves.(14) Women were a little less likely to steal items valued at less than 10s. than were men, a little more likely to take things valued from 10s. to [pounds]10, and then, finally, less likely to steal items worth more than [pounds]10. In all, two-thirds of both men and women pilfered goods valued up to 40s.(15)

There were, however, marked differences with respect to what each sex stole. A preliminary analysis of 1780, 1785 and 1789 reveals very similar patterns for all three years. As Table Two shows, clothes and household goods were the most likely targets of theft for both men and women, but disproportionately so for women. Men, on the other hand, were more likely to take industrial materials, livestock, horses, tools and, perhaps surprisingly, food. Much of the food consisted of items like tea or sugar, which were stolen from quays, wharves and warehouses where the men worked or claimed to be looking for employment. The tools and industrial materials also tended to come from places of employment (though the accused was often not an employee), from buildings under construction or from storage sheds or stables. The livestock frequently disappeared from London's markets or on their way to such, while horses were taken from inns, stables and fields.

[TABULAR DATA FOR TABLE 2 OMITTED]

These were not places in which women and girls had cause to be in the normal run of events. As noted above, female thieves were particularly likely to take clothes and household items. Clothes were a valuable commodity, constituting the largest category of household expenditure.(16) Moreover, they could easily be pawned or sold in the extensive second-hand clothing market.(17) Women not only had ready access to both clothing and household goods, but they would have had an intimate knowledge of the value of such items, given their family responsibilities. Aside from clothing and household goods, women also tended to take cloth (often shoplifted - and this activity was largely a female undertaking). As well, women were more likely than men to take money, watches and jewellery. This resulted, for the most part, from the actions of prostitutes who were frequently accused of stealing from their clients.

These appropriation patterns resulted, in large measure, from the opportunities males and females experienced in their normal range of activities. This was also the case with respect to the reasons the accused women and men gave for their actions. In reading through the OBSP, the plethora of explanations and justifications offered by the accused in their defence is striking. It is possible to group these, however, into eight categories (please refer to Appendix A for a summary of these).(18) The first is made up of cases in which the accused either made no defence statement or in doing so, said nothing with respect to his or her guilt or innocence. In cases where the evidence was deficient or the prosecutor or a witness did not appear, those accused were not usually put on their defence. As indicated earlier, a number of cases were also recorded very cursorily, and the percentages of such cases increased in the years when the numbers were high. When the accused did speak - -or when this statement was recorded - they sometimes said nothing more than 'I have character witnesses' or 'I have nothing to say in my defence' or alternatively, 'I leave my defence to my counsel.'

The second group of reasons consisted of variations of 'I had the stolen item, but not through theft'. Claims to have bought, found or borrowed the item in question were common as was the insistence that 'I was hired to deliver it', or that 'it was given or loaned to me', or again 'I did not intend to steal it.' The third group of reasons threw doubt on the upright character of the prosecutor or on the veracity of the prosecutor's story. Claims of malicious prosecution, lying witnesses or prosecutors being too drunk to recall the incident were other common explanations in this group. Fourth, were the claims of innocence with no further justifications added to them. Fifth, were the reasons the accused believed ought to excuse him or her from accepting responsibility for the theft. Most common was the claim that the accused had been too drunk to remember what had happened. The sixth group of reasons saw the accused trying to portray themselves as the victims of circumstance. There were innumerable variations on the claim 'I was just coming along and was taken up'. Some of those accused claimed that someone else took the item, or at least had the opportunity of doing so. Other defendants claimed to have been at the place where the theft occurred for a legitimate reason. Seventh, were those who admitted their guilt; some cited distress as the cause while others begged for mercy. Finally, there were usually a few cases which simply could not be classified or in which the meaning of the statement by the accused was unclear.

The defence statements have been analysed for the decade as a whole.(19) It is apparent (according to Table Three) that Group Two ('I had it, but not through theft') was the largest category for both males and females. For males, Groups Six (the victims of circumstance), One (those who said nothing concerning their guilt) and Four (those who claimed innocence) came next and were all fairly close to being about one-half the size of Group Two. For females, Groups Four, One, Six and Three (those who disparaged the prosecutor or his or her story) were ranked next.

Females were also more likely than males to plead guilty, although for both sexes the number doing so was quite small, no doubt in part because the court at times actively dissuaded defendants from doing so.(20) Even fewer people claimed that they should not be held responsible for their actions, though males clearly outnumbered females here. This was chiefly due to claims of being excessively drunk. Females may have been more willing to plead guilty in the hope that they would receive more lenient treatment by virtue of their being women - in effect, playing on the prejudices of the time. These women tried to present themselves as vulnerable - a few made a point of telling the court they were pregnant,(21) for instance, while others claimed responsibility for children.(22) Whether or not this was a deliberate attempt to manipulate gender prejudices, presenting oneself as vulnerable was a strategy which benefitted women far more than men. Women were almost twice as likely to be found not guilty or to have their sentence reduced as were men who pleaded distress or who asked for mercy. The figures for those who benefitted from pleading mercy in the decade were 39 per cent for females and 21 per cent for males. For distress, the figures were 44 per cent for women and 24 per cent for men.

Turning from simple assertions of guilt or innocence to the explanations the accused presented in their defence, there are again some marked differences between those offered by males and by females, particularly with respect to Group Six. Men were more likely to claim to have been the victim of circumstances than ere women. The question, of course, is why. If the group is broken down into its four components - 'someone else took it,' 'I was there for a legitimate reason,' 'I was coming along and was taken' and lastly, other reasons - it is apparent that the gap between males and females comes wholly from the third sub-group.

Table 3

MOTIVES FOR THE DECADE

 Males (%) Females (%)

Group I(#) 542 (16.4) 184 (14.5)
Group II 1188 (35.9) 436 (34.2)
Group III 200 (6.0) 142 (11.2)
Group IV 479 (14.5) 214 (16.8)
Group V 81 (2.5) 27 (2.1)
Group VI 620 (18.7) 158 (12.2)
Group VII 125 (3.8) 75 (5.9)
Group VIII 77 (2.3) 40 (3.1)

 3312 1276

* percentage of males or females

# I have omitted those for whom no defence was recorded; that is,
2125 males and 782 females. Hence, the overall totals are lower than
those given in Table One.

There was one other group with marked differences in the stories males and females told. As a whole, the percentages of males and females who claimed they had the stolen item in their possession for legitimate reasons was very close - just over one percentage point separated them. If Group Two's components are examined, however, a striking difference emerges. Males dominated the first three sub-groups ('I was hired to deliver it."I found it' and 'I bought it'). Just over one-quarter - 26.2 per cent - of all males who made a defence statement fell into these three sub-groups. Only 12 per cent of women and girls making a defence statement cited these reasons. In two other sub-groups, 'I pawned it for another or with permission,' and 'it was loaned or given to me', the situation is reversed: females dominate. Women and girls in these two sub-groups constituted just over 14 per cent of all females who made defence statements. Men and boys, on the other hand, made up just less than four per cent of males giving such statements. Again, one is faced with the need to try to explain the polarisation within this group.

There is an explanation for this situation and for the male dominance of the one sub-group in Group Six ('I was coming along and was taken') which can be inferred from a pattern in this behaviour. The people who claimed they had been hired to deliver the item almost always said they had been so by a stranger (who usually disappeared almost immediately thereafter). Similarly, they usually said the items were purchased from a stranger, or someone known slightly who could not now be found. Those who claimed they were simply coming along, almost always said they were accused by strangers. Finally, those who said they found the items, claimed no relationship whatever with the prosecutor or other people involved in the incident.

In the sub-groups dominated by women and girls, however, there was in most instances, a relationship with the prosecutor and others involved in the apparent theft. Those accused usually said they knew the person who had given them the item or from whom they had borrowed it. In a number of pawning cases, it is apparent that these were instances of women's borrowing networks gone wrong - that is, one partner had been a little too free with the belongings of another.

While overall, the sub-groups being discussed constituted a minority of the thefts perpetrated by males and females, this pattern - that the latter were more likely to claim to know their prosecutors and others involved - also holds overall. If the cases are examined in which it is clear that the prosecutor and the accused knew one another, females again dominate. The number of cases in which it was possible to identify this state of affairs always constituted a minority of the overall total. In part, this resulted from the cases in which no defence statement was made. It was also often unclear from the internal evidence whether the accused and the prosecutor knew each other. The following figures from Table Four, then, most certainly under-estimate the actual situation. Five years have been examined: 1780, 1782, 1784, 1786 and 1788. In each of these five years, a substantially greater proportion of women and girls knew their prosecutors than did men and boys. Only once, in 1780, was the spread between males and females less than 14 points. In that year, 21.6 percent of men and 28.7 percent of women knew their prosecutor. In 1782 and 1786, the spread was almost 14 points - the figures for males and females respectively being 21.2 percent and 34.6 percent in 1782, and 19 percent and 32.8 percent in 1786. In 1784, 19.5 percent of males and 42.5 percent of females knew their prosecutors, while in 1788, the figures rose to 25.2 percent for males and 55.1 percent for females. In most years, moreover, at least half of the men and women who knew their prosecutors were servants.(23) Women in 1784 and 1786 were the exceptions to this. In every year but 1788, when there was a near tie, male servants formed a larger percentage of those knowing their prosecutor than did female servants. This means that the gap between males and females living in the community (rather than with their employers) was even greater than the figures in Table Four indicate. The patterns from the sub-groups, then, are borne out in these overall figures.

What is the meaning of this pattern? Why did women steal more frequently from those whom they knew? I think, in large measure, this was because the geographic boundaries of quotidian life were more circumscribed for women than for men. Most women married, and hence, were responsible for household and children. To be sure, these women were not confined to the home; Victorian notions of separate spheres did not obtain at this time. Plebeian women made frequent forays into the streets - to fetch water from cisterns and local taps, to buy food and to gossip at the local shop (where most of them were sufficiently well known to run credit tabs), to lend and borrow items to and from neighbours, and to drink at the local pub. As the statements of witnesses in defamation cases before the consistory courts reveal, life was very public in this period, and neighbours very watchful.(24) London was not a city of anonymous, isolated individuals; people were known within their neighbourhoods, and women, in their daily lives, had less cause to leave those neighbourhoods than did men. Many of the most common occupations in London required men to move beyond the immediate vicinity where they lived. Porters on delivery, dock workers, those engaged in the building trades, small masters delivering shoes, clothing or furniture to often distant middlemen all regularly moved beyond their neighbourhoods. The logic of this situation clearly suggests that women's opportunities for theft occurred within their own neighbourhoods, and consequently, they were more likely to involve the goods of someone they knew.

Table 4

THOSE WHO KNEW THEIR PROSECUTORS

 Males Females

1780 41 (21.6)(*) 48 (28.7)
1782 50 (21.2) 55 (34.6)
1784 78 (19.5) 45 (42.5)
1786 82 (19.0) 39 (32.8)
1788 78 (25.2) 49 (55.1)

* percentage of males or females

While geographic constraints are an important reason accounting for the female predilection to steal from those whom they knew, their behaviour also resulted from the fact that females were more likely to engage in certain forms of mutuality - in particular, borrowing networks. These networks were an important makeshift for plebeian women trying to make ends meet, and these women, subject as they were to lower wages, quite commonly loaned each other items to pawn rather than money. Cases involving this kind of behaviour surfaced in the Old Bailey Sessions Papers. In the 1780s, just about seven per cent of females who made a defence statement claimed that they pawned the stolen item for another, that they did so with permission, that they did so to settle a debt or with the intention of returning it. For males, the figure was only 0.8 per cent. Some of the women who had been involved in pawning, however, gave other reasons for their actions in their defence statements.(25) In all, the pawning of goods for the above reasons figured in about 15 per cent of the cases of female theft and two per cent of male.(26) While borrowing networks did not always involve pawning, and so in reality would have been considerably more extensive than the OBSP material indicates, the pawning cases do provide a window onto this informal behaviour.

The preponderance of women pawners was not a peculiarity of the OBSP, however. That women were much more likely to take part in borrowing networks than men is well known thanks to the work of Deborah Valenze and Ruth Smith(27) for the early nineteenth century and Ellen Ross for London later in the century.(28) I am not claiming that women practised mutuality while men didn't. Anyone who has examined labouring-class autobiographies knows that plebeian men, too, had a collective ethos.(29) The mutuality they practised, however, took different forms from that of women. Men were more likely to help each other find work in slack periods, for instance. Where females' mutuality tended to be neighbourhood based, males' tended to be work centred.

Because borrowing networks were informal, and little direct evidence has survived concerning their nature and their limitations, the OBSP cases constitute one of the few sources we have for this behaviour. Since the dynamics of these networks have rarely been explored, it is worth pausing to consider their characteristics and the ways in which they could go wrong. A number of borrowing network features can be teased out from the OBSP cases. In the first place, the networks were ongoing. In February 1780, for instance, Catherine Amer accused her roommate, Ann Friend, of stealing four pairs of silk stockings. During the trial it came out that Amer had previously borrowed and pawned two neckcloths from Friend (the latter claimed she had pawned the stockings in order to get her neckcloths out of pawn). In June of the same year, Ann Powell was accused of stealing clothes and other things from her sister's landlady. In her defence, Powell stated, "I used to do anything she [the prosecutor] wanted; when she wanted money to pay her washerwomen she sent me to pawn things for her; she knows she gave me orders to do it to bring money to her."(30)

Borrowing network transactions could become convoluted: in September 1782, for instance, Sarah Skettles bought a waistcoat and some aprons from a cloathesman for 10/6. She borrowed part of the money from a neighbour and then pawned one of the aprons in order to repay her.(31) Two years later, Martha Ray claimed her prosecutor, Catherine McCue, had borrowed half a guinea from her. Ray said that when she asked her for it, McCue "told me she could not pay me, and she gave me the cloak with her own hands (which she was now accused of stealing) and I went and pawned it for my own half guinea."(32) The women in both these cases were engaged in a precarious, complex juggling act, a precariousness, moreover, which was the common experience of plebeian life generally. Certainly, this precariousness is reflected in the fact that money was not always the basic currency regulating the exchange. Rather, at times it became another commodity - being purchased and sold by the exchange of clothes or other items at hand. The exigencies of this situation, moreover, required a certain canniness on the part of plebeian women. This was apparent in the 1781 prosecution of Ann Braidy. She asked Hannah Smith to pawn some curtains and two pairs of silk stockings for her. Braidy did so because she was "rather dirty;"(33) Smith being cleaner would presumably look less needy and hence, would be able to coax a better price out of the pawnbroker.

The borrowing networks were meant to meet short-term difficulties. Typically, no more than a few days passed before the giver expected the loan to be returned somehow. Thus, when Margaret Rowe claimed to have found some cotton in 1782, her landlady, a Mrs. Mason, offered to buy it, but had not the money. According to Rowe, they agreed that she [Rowe] would pawn it until the end of the week, when Mason expected money with which she would pay Rowe the difference between the pawning and the real value.(34) In 1785, Elizabeth Bland told the court that her prosecutor had loaned her the clothes (which she was now accused of stealing) in order to pawn them, and had given her one week to get them back.(35)

The on-going nature of the borrowings, the precariousness and complexity of the transactions and the need for short-term satisfaction in these networks could lead to misunderstandings resulting in theft charges. While plebeian women often behaved with generosity to one another, it is clear that lending and borrowing were transactions in the business of making ends meet just as was waged employment. At times misunderstandings between the network partners seem to have been the product of a somewhat fluid notion of private property. Both women and men seemed to feel that they had the right to use the property of others, especially when in distress. Throughout the decade there were instances of people - usually women - pawning items from the furnished rooms they rented. On average, nine women and four men a year admitted doing so.

Those accused usually admitted they had made use of the item (by pawning it for cash), but often claimed that they had been in distress and that they had intended to redeem it. Some commented resentfully that they would have made good had they not been taken up. Ann Mitchell, in 1782, for instance, pointed out that she had pawned the things repeatedly and always returned them.(36) In 1787, Elizabeth Gosling told the court she had already returned some of the things, and in any case had not left the lodging.(37) On occasion, even the prosecutor would admit that the accused had not meant to steal the things. In 1785, Elizabeth Cooper's landlady made such an admission and Cooper was found not guilty.(38) The things were often pawned when the tenant was in distress or when an extraordinary demand for money came up. In 1788, Jane Williams' landlord demanded a week's rent in advance when she took the room. To meet this she pawned the bedding, a shovel and various other things in the room. Before she could redeem the items, the landlord missed the things and had her arrested.(39) Both women and men pawned such goods, but women did so more frequently. They did seem to believe they had the right to use the items in the rooms they had rented. While the numbers cited in the OBSP are small, it is very likely that they would have been substantially higher had the squibs been more complete. Certainly, the alacrity with which landlords pressed charges suggests the pawning of such goods was a widespread problem. These OBSP cases, then, offer a small window onto a different understanding of property, one which was more fluid than that being put forward by the upper classes generally. And as to why women were dominant, I think this is an instance where their greater geographical circumscription may have made them more willing to resort to this makeshift. Since women did most of the pawning at this time,(40) and were more likely to be part of borrowing networks, using the goods in their rented rooms may have seemed a sensible and practical response to distress.

This somewhat fluid notion of private property was not limited to goods from rented rooms, but is apparent in other kinds of borrowing as well. In the most extreme instance, the woman charged with theft would claim that the prosecutor had told her to use whatever she wished. Thus, Mary Robinson in 1781 told the court that Jane Stewart had said, "I was welcome to stay with her, and was welcome to anything she had."(41) Similarly, in 1785, Catherine Knock claimed that the prosecutor's wife had said she could take anything to pawn, and for good measure added that this woman had also allowed her to wear her clothes.(42) In these various cases, those accused assumed, erroneously as it turned out, that they had the right to use the items to meet short-term needs.

Fluid notions of private property were not the only source of difficulties, however. As discussed earlier, lenders often needed to be repaid (in whatever fashion) within a fairly short period of time. If the borrower exceeded the length of time the lender thought proper, then, as Elizabeth Bland discovered above, theft charges could result. Other problems could also land a borrower in court: choosing to repay a debt in a manner unacceptable to the lender, for instance. Ann Wood, in 1780, claimed to have left other clothes to replace those that she had pawned. In this instance, the prosecutor denied she had done so and Wood was found guilty.(43) The complexities of the networks caused problems as well. At times, third parties claimed that the lender had, in fact, stolen the item on loan, with the result that the borrower could find herself up on theft charges. Mary Williams claimed to be just such a victim in 1781(44) and Susanna Kelly in the following year insisted she had no idea where Mary Walker got the items she had given her.(45)

Yet another way in which the network could go wrong resulted from the relationship between the partners. If there was a falling out between them, the borrower might find herself charged with theft (whatever the original agreement had been). Thus, Elizabeth Green in 1781 claimed that Jane Barber had loaned her a gown the latter had been altering for Catherine Geary. Barber and Green then went out together and when they ran out of money Barber told Green to pawn the gown. Shortly thereafter, the two "had some words" and Green claimed that Barber had had her arrested out of spite.(46) In 1785, Jane Curtey said she had her landlady's permission to pawn the bedding since she was in distress. Subsequently, the two had words and Curtey claimed she was taken up as a result.(47)

Finally, some women charged with theft thought to take advantage of the common knowledge that such borrowing was widespread. Thus, in 1781 Elizabeth Jones claimed that an unnamed young woman had asked her to pawn the coat she was accused of stealing on the promise of 6d. and a pot of purl. The court did not believe her.(48) While in this instance, the claim did not seem very plausible, in other cases the courts were more willing to accept these justifications, especially if there was evidence of an ongoing borrowing relationship between the accused and the prosecutor. Again in 1781, Mary Hughes claimed that her mistress, Elizabeth Glover, had asked her to pawn the clothes and bedding she was accused of stealing. When the court learned from the pawnbroker that Glover had sent Hughes before to pawn things, a verdict of not guilty was returned.(49) Similarly, in 1788, Jenny Mead claimed often to have pawned things for her prosecutor. When the latter admitted to having had Mead pawn things in past, the court refused to convict.

Borrowing networks, then, could go wrong in a number of ways. Hans Medick has pointed out that social exchange made sound economic sense for plebeian women and men during this period since "it produced or reproduced just that solidarity to which small producers could, in times of dearth, crisis and need, most easily have recourse."(50) In the borrowing networks, however, social exchange did not simply promote and support access to economic assistance. Rather, the two became inextricably intertwined and the very characteristics, not to mention the informality, which made borrowing networks most useful to their female practitioners could also lead to misunderstandings and in some cases to legal action.

To sum up, then, women were more likely to be charged with stealing from those whom they knew because they were more bound to their neighbourhoods than were men, and because borrowing networks - part of the ethos of mutuality whose forms were themselves gendered - could go wrong in a number of ways. There is evidence as well that while both women and men had a less than rigid notion of private property, in the case of women especially, this could result in their landlords and their borrowing partners pressing theft charges.

Garthine Walker's claim that the dynamics of female theft offer a more complex view of gendered experience than many historians of crime have hitherto acknowledged is certainly borne out by the OBSP material. While need was a major determinant in the decision to steal for both men and women, there were decided differences between male and female thieves: the latter stole different kinds of things than did men and the reasons they offered for their actions reflected the different patterns and rhythms of female life. Female thieves were not simply paler reflections of the male norm. Indeed, the differences may run even deeper than this paper suggests. The dramatic increase in the 1780s in the kinds of theft charges under consideration was shown in Table One. For males and females combined, the figures ranged from 441 in 1780 to a high of 954 in 1785. What is striking, however, is the fact that males account for all of this increase. Prosecutions against women, in fact, fell by the end of the decade.(51) Why was this the case when women were more economically vulnerable than men? John Beattie has shown that between 1660 and 1800 the number of prosecutions against female thieves did fluctuate with economic conditions(52) - as one would expect. Douglas Hay has found similar patterns with respect to Staffordshire in the second half of the eighteenth century. So why in the difficult decade of the 1780s, did this relationship not hold in London? Were the authorities and prosecutors acting on the belief that was widely held in the eighteenth century: that after a war, demobilised men caused a crime wave? Were they, in fact, targeting male perpetrators for prosecution? Does this explain why female figures fell after 1783? Or were women resorting to other makeshifts not used by males? In the workhouse of St. Martin in the Fields, one of the largest in London, women always made up two-thirds to three-quarters of the house population in the late eighteenth and early nineteenth centuries.(53) Similarly, they constituted a clear majority of the clients of what came to be the Mendicity Society, a large London charity of the period.(54) Do poor relief and charity or the recourse to prostitution explain why fewer women resorted to theft? Or did the female figures result from a combination of these factors? Or even from other reasons entirely? Obviously, much more work needs to be done in order to understand this situation. It is clear, however, that female theft patterns, in this decade at least, were moving according to very different rhythms from those of males. Whatever the answer to this last riddle, it is apparent from this study that there were significant differences particularly with respect to the justifications men and women accused of theft offered for their actions. These justifications, moreover, emanated from the opportunities and circumscriptions of quotidian life for each sex. In short, it is necessary to understand theft as a gendered experience.

Department of History Brandon, Manitoba, Canada R7A 4T5

APPENDIX A

BREAKDOWN OF DEFENCE STATEMENTS

GROUP ONE: NOTHING SAID CONCERNING GUILT OR INNOCENCE

No defence recorded. I leave my defence to my counsel. Other reasons (1 have character witnesses, etc.).

GROUP TWO: I HAD IT BUT NOT THROUGH THEFT

I was hired or asked to deliver it. I found it or I bought it. I pawned it for another or with permission or I meant to redeem it. It was loaned or given to me. Other reasons (I didn't intend to steal it, etc.).

GROUP THREE: THE PROSECUTOR OR HIS OR HER STORY ARE SUSPECT

The prosecutor was too drunk to know what happened. Other reasons (the prosecutor or witness are lying, etc.).

GROUP FOUR: I AM NOT GUILTY

I am innocent or I know nothing of it. I can prove I did not take it. Other reasons (I wasn't with the other defendant, etc.).

GROUP FIVE: I AM NOT RESPONSIBLE FOR MY ACTIONS

I was too drunk to remember what happened or to know what I was doing. Other reasons (I am subject to fits, etc.).

GROUP SIX: I AM THE VICTIM OF CIRCUMSTANCES

Someone else took it or had the opportunity of doing so. I was there for a legitimate reason. I was just coming along and was taken up. Other reasons (The prosecutor and I were drinking and he/she accused me, etc.).

GROUP SEVEN: I AM GUILTY

I did it from distress. I beg mercy. Other reasons (I only took some of the things, etc.).

GROUP EIGHT: NOT CLASSIFIED

Miscellaneous. Unclear

ENDNOTES

I would like to thank John Markoff for his commentary on an earlier version of this article given as a paper at the North American Conference on British Studies in Chicago on 18 October 1996. I am also grateful to Douglas Hay for commenting on the draft of this piece. Any errors that remain are, of course, my own.

1. John Beattie, " 'Hard Pressed to Make Ends Meet': Women and Crime in Augustan London," Women and History, Valerie Firth, ed. (Toronto, 1995), p. 106.

2. Garthine Walker, "Women, theft and the world of stolen goods," Women, Crime and the Courts, Garthine Walker and Jenny Kermode, ed. (Chapel Hill, 1994), p. 99.

3. For a discussion of the reformers of this period, see Michael Ignatieff, A Just Measure of Pain (New York, 1978), Ch. 3, pp. 44-79.

4. T.S. Ashton, Economic Fluctuations in England 1700-1800 (Oxford, 1959), pp. 162-169.

5. Ashton, p.24.

6. L.D. Schwarz, "The Standard of Living in the Long Run: London 1700-1860," Economic History Review, Second Series, Vol. XXXVIII, No. 1 (February 1985): 31.

7. Old Bailey Sessions Papers, British Library, PP 349a 30, 1779-1789.

8. So when I refer to 1780, for instance, I am including the December 1779 session through to that held in October 1780 - the last in the legal year.

9. John Langbein, "Shaping the Eighteenth-Century Criminal Trial: A View from the Ryder Sources," University of Chicago Law Review, Vol. 50, No. 1 (Winter 1983): 14. The squibs as a percentage of total male theft cases for the decade are as follows: 1780, 14.7%; 1781, 27.6%; 1782, 32.3%; 1783, 37.4%; 1784, 46.1%; 1785, 36.3%; 1786, 38.1%; 1787, 44.8%; 1788, 44.7% and 1789, 34.7%. For women the respective figures are 18.3%, 26.3%, 28.1%, 34.2%, 51.4%, 40.1%, 44.2%, 44.2%, 50.9% and 31.6%.

10. A potentially serious problem arises from the squibs. If they target one group disproportionately - women who stole items of low value, for instance - then the analysis of this paper would obviously be distorted. It is necessary, then, to identify the defendants whose cases were not fully reported. In examining the squibs over the course of the decade, it is apparent that no one group was being disproportionately targetted. If these ten years are averaged, the squibs account for 37.9 per cent of male thieves and 37 per cent of female. There was considerable variance within the decade, but only in two years did the difference between male and female squibs reach six per cent: in 1786 when females were more numerous, and in 1788 when the converse was true. Moreover, an analysis of the squibs in 1780, 1782, 1784, 1786 and 1788 indicates that most of these cases were minor. A little less than three-quarters of them involved the theft of goods valued at 40s. or less, a pattern which was a little more pronounced than that for all thieves in this decade. As will be seen below, two-thirds of those accused of theft in the 1780s stole goods valued at 40s. or less. Once again, there was almost no difference between male and female squib patterns: 71.4 per cent of male squibs stole goods valued at 40s. or less, while 71.2 per cent of female counterparts did so. The squibs, then, do not significantly distort the analysis of theft in the 1780s.

11. Recording practices changed beginning in 1782. Where previously each individual had been assigned a separate case number (even if several people were named on one indictment), in 1782 case numbers began to be assigned to the indictment rather than to the individuals involved. Thus, if several people were named in one charge, from 1782 onward they were usually all assigned the same number. Given this situation, either the charges or the individuals named in them can be counted throughout the decade. Since I am primarily interested in the behaviour of men and women, I've opted for the latter. Individuals up on the same charge often offered different defences and I wished to attend to this. The figures have not been corrected for repeat offenders since they aren't always identified in the yearly indices or in the cases themselves. Much of the time they are identified, but not always.

12. See Walker, pp.83-87.

13. One way in which females in the 1780s differed from those in Walker's study was their greater propensity when acting in collusion with others, to act with males. In the 1780s, females accused of theft were more than twice as likely to act in consort with males than were the latter to act with women. The percentages, however, were small - three per cent for males and 7.1 per cent for females - which would support Walker's overall claim that males and females rarely acted with members of the opposite sex.

14. In considering the value of the items stolen by males and females, there is an immediate problem. The valuations contained in the lists of items taken are sometimes missing or incomplete. For instance, for 2.4 per cent of the males and 0.4 per cent of the females during the decade no value was given for the items stolen. The value of the goods may also have been deliberately underestimated since a number of charges became capital once goods of a certain value were stolen. No doubt there was also a lot of guesswork in trying to put a value on various small household items or on industrial bits and bobs. In spite of these difficulties, John Beattie has convincingly argued that without claiming great accuracy for these valuations, they do nevertheless indicate "in an approximate way the general levels of prosecuted theft" (See John Beattie, Crime and the Courts in England [Princeton, 1986] p.183). I am, in any case, less concerned with the precise real value of the goods stolen than with a comparison of male and female patterns.

15. Walker also found this to be the case, although in her study she considered only grand larceny and burglary.

16. Walker, p.89.

17. For a discussion of the second-hand clothing market, see Beverley Lemire, "Consumerism in Pre-industrial and Early Industrial England: the Trade in Secondhand Clothes," Journal of British Studies, Vol. 27 (January 1988): 1-24.

18. Cynthia Herrup has discussed the defences offered by those accused in the seventeenth century in The Common Peace (Cambridge, 1987), pp.147-148. The tripartite scheme she developed (claims that the goods had been legitimately obtained, that they had been found and that the defendant knew nothing of them) is too broad for my purposes.

19. Before discussing motivation, however, there is a caveat to be considered. The men and women and girls and boys who were accused of theft had a lot to lose - in some cases their lives - should the court believe the prosecutor's version of events. Obviously, the defence statements were tailored to present the accused in the most positive light. This being so, they cannot simply be taken at face value - virtually no one admitted to being a professional thief, for instance, though clearly some were. Nevertheless, it was necessary that defence statements be plausible, that they contain at least some truth when reinterpreting events. It was also necessary that they offer explanations which reflected the realities and rhythms of London plebeian life. The actions described in the statements had to seem to be likely behaviour. So while their strategic purpose must be borne in mind, the statements do give a general picture of the most plausible justifications on offer.

20. See, for instance, OBSP, December 1782, Case 5, William Smith.

21. See, for instance, OBSP, December 1781, Case 7, Mary Carpenter; or January 1788, Case 119, Jane Williams. Claims of pregnancy could also be used to postpone capital punishment.

22. See, for instance, OBSP, September 1780, Case 479, Mary Collins; or February 1783, Case 298, Ann Webb; or again, September 1784, Case 854, Sarah Hull.

23. For the years 1780, 1782, 1784, 1786 and 1788 respectively the percentage of male servants accused of theft who knew their prosecutor was as follows: 53.7%, 60%, 55%, 51.2% and 53.9%. For female servants the corresponding percentages were 47.9%, 55%, 37%, 35.9% and 55.1%.

24. See Peter Earle, A City Full of People (London, 1994) for an extended discussion of this for London in the period 1650 to 1750. Little had changed thirty years later as an examination of the defamation cases from 1780 to 1820 reveals. See also the Consistory Court Depositions, Greater London Record Office, 1780-1820.

25. That is, the pawning was mentioned in passing while some other reason was cited as the main cause of the apparent theft.

26. This claim is based on a close reading of the cases in which the accused gave defence statements for the five years from 1780 to 1784. The cases making up the percentages include those where either the accused or a witness indicated that the goods had been pawned with permission, or that they had been loaned in order to be pawned, or that the accused had intended to redeem them, or finally, that the accused had pawned the goods in order to settle an outstanding debt (usually wages which the prosecutor had failed to pay). The actual percentages of women in 1780, 1781, 1782, 1783 and 1784 respectively were as follows: 19.2%, 16.4%, 12%, 10.6% and 15.1%. For men, they were 2.1%, 1.3%, 2.5%, 2% and 2.5%.

27. Ruth Smith and Deborah. Valenze, "Mutuality and Marginality: Liberal Moral Theory and Working-Class Women in Nineteenth-Century England," Signs, Vol. 13, No. 2 (1988): 277-298.

28. Ellen Ross, "Survival Networks: Women's Neighbourhood Sharing in London Before World War One," History Workshop Journal, No. 15 (Spring, 1983): 4-27, and Love and Toil (New York, 1993). The makeshifts Ross describes were also common in the late eighteenth century. Indeed, it may be that the establishment of borrowing networks in the face of economic uncertainty, is a fairly standard response. Certainly, Carol Stack has shown that similar behaviour (if on a somewhat more extended basis) existed among Afro-American women whom she studied in All Our Kin (New York, 1974), pp.32-44. My thanks to John Markoff for alerting me to this reference.

29. See Pat Hudson and Lynette Hunter, ed. "The Autobiography of William Hart, Cooper, 1776-1857, A Respectable Artisan in the Industrial Revolution," London Journal, Vol. 7, No. 2 (1981), Part 1, Also see Mary Thale, ed., Autobiography of Francis Place (Cambridge, 1972).

30. OBSP, February 1780, Case 127, Ann Friend and June 1780, Case 341, Ann Powell.

31. OBSP, September 1782, Case 516, Sarah Skettles.

32. OBSP, January 1784, Case 155, Martha Ray.

33. OBSP, January 1781, Case 104, Ann Braidy.

34. OBSP, September 1782, Case 504, Margaret Rowe.

35. OBSP, September 1785, Case 761, Elizabeth Bland.

36. OBSP, February 1782, Case 276, Ann Mitchell.

37. OBSP, September 1787, Case 859, Elizabeth Gosling.

38. OBSP, December 1785, Case 112, Elizabeth Cooper.

39. OBSP, January 1788, Case 119, Jane Williams.

40. Melanie Tebbutt, Making Ends Meet (New York, 1983), Ch. 2, pp.37-67. Tebbutt's findings for the late nineteenth and early twentieth centuries also held for the late eighteenth century. See my dissertation "Myths of Power: the Knowing of Poverty in London, 1790-1815" (York University, 1991).

41. OBSP, May 1781, Case 336, Mary Robinson.

42. OBSP, June 1785, Case 705, Catherine Knock.

43. OBSP, April 1780, Case 179, Ann Wood.

44. OBSP, April 1781, Case 198, Mary Williams.

45. OBSP, December 1782, Case 67, Susanna Kelly.

46. OBSP, September 1781, Case 565, Elizabeth Green.

47. OBSP, June 1785, Case 642, Jane Curtey.

48. OBSP, January 1781, Case 10, Elizabeth Jones.

49. OBSP, April 1781, Case 204, Mary Hughes.

50. Hans Medick, "Plebeian Culture in the Transition to Capitalism," Culture, Ideology and Politics, Raphael Samuel and Gareth Stedman Jones, ed. (London, 1982), p.92.

51. The question, of course, is what the increase actually measures - whether the figures represent an actual increase in appropriation, that is, instances of theft (and whether this is simply a reflection of population growth), or whether the figures measure changes in what has often been called 'control' - the behaviour of authorities. It is unclear that this question can ever be definitively answered; certainly it cannot on the basis of the OBSP material. It may be that the increase measures a combination of these various factors. Douglas Hay has shown in "War, Dearth and Theft in the Eighteenth Century: the Record of the English Courts," Past and Present, No. 95, (May 1982), that, as contemporaries believed, after every war in the eighteenth century the amounts of theft did increase dramatically, thus delineating a clear relationship between historical eventuation and indictment levels. The economic problems of the 1780s also make it very probable that the increase does more or less reflect actual appropriations. On the other hand, it is difficult not to believe that faced with continually increasing numbers of charges (whatever their prime cause), the authorities and prosecutors did not act more stringently.

52. John Beattie, "The Criminality of Women in Eighteenth-Century England," Journal of Social History, Vol. VIII (Summer 1975): 102-103. A shorter discussion of the same issue is to be found in his Crime and the Courts in England, pp.242-243.

53. See my dissertation, "Myths of Power: the Knowing of Poverty in London, 1790-1815," and my article, "A Culture of Poverty? The St. Martin in the Fields Workhouse, 1817," Journal of Interdisciplinary History, Vol. XXVI, No. 2 (Autumn, 1995).

54. Between 1800 and 1803 Matthew Martin interviewed 2,000 mendicants, of whom 90 per cent were women. Matthew Martin, Letter to the Rt. Hon. Lord Pelham on the State of Mendicity in the Metropolis (London, 1803). The Society also investigated begging letters, some 85 of which are to be found in the papers of the second Earl Spencer from the years 1825 and 1826. Of these, 56 or 65.9 per cent were from women. British Library, Manuscripts Room, Althorp Papers, G 141 and G 145.