In October of 1620, the family of Thomas Conant–a husbandman of Moretonhampstead on the edge of Dartmoor, Devon–fell under suspicion of stealing sheep from their neighbour John Hutchins.Footnote ¹ Curiously, Thomas himself was not examined by the magistrate, or if he was his testimony does not survive, but other members of his household were. His wife Joan was asked to explain the presence of two ewe sheep that the local constable had found in her husband’s barn with their legs bound by a piece of cord. Joan denied any knowledge of the sheep-napping, but she explained of the ‘60 weight of wool’ also found in their house by the constable that ‘she bought the same of certain men of Chagford [4 miles away] whose names she knoweth not’. Her daughter Thomasine was examined next, and the magistrate asked her if a basket of mutton found in the barn was her mother’s: she replied that it could not be, as ‘their basket was sent to the house of John Howe of St Thomas [a suburb of Exeter 12 miles away] unto a sister of this examinant about three weeks before Michaelmas last to fetch home this examinants clothes’, and that the said basket ‘was either left there or lost by the way’. The magistrate then examined John Conant, a son of Thomas and Joan who had ‘served Giles Kingwell of Crediton [a cloth town 12 miles away] … by the space of five years until Lammas last and then he came home unto his father’s house’. He too denied all knowledge of any sheep stealing but did confess that ‘he was on the said Hutchins’ ground on Saturday last and did drive a mare into the said ground’.

His brother Benjamin was the next examinant to face the magistrate and in a more fulsome response than his siblings described a plethora of work tasks. Where had his sister been on the Saturday that the sheep went missing? She was at home in bed sick all day ‘until it was almost night, at which time his sister rose to bake three loaves of bread’. What about his brother? ‘John went forth in the evening into their own ground to set or till springles [traps] to catch woodcocks.’ What did he know about the basket? He himself had ‘carried the basket unto the aforesaid Howe’s house to fetch peas from the said house, but there were none, and therefore left the basket there’. The constable had also found a basket of wet wool when searching the Conant house and barn, but here too Benjamin Conant could explain: ‘it was wool he gathered up about the house, and washed the same himself to make him clothes’. And whilst he also denied any involvement with the sheep theft, he did confess that the family owned a length of cord, for ‘his brother Richard brought with him from sea about ten days last past a certain bunch of cord, which this examinant sayeth his mother did use to hang clothes to dry from’.

It is not possible to know if the Conants took the sheep, or whether the various work tasks they attested to were cover stories or accurate accounts of their everyday work tasks. But if we accept that the stories the family told to the magistrate were plausible accounts of everyday work activity, and therefore reflect the realities of everyday life in this society, we can draw from the case a number of insights that are crucial to understanding the experience of work in early modern England. For a start, it provides a reminder that household economies in rural England were remarkably versatile and diverse: there is evidence here of engagement in commerce, baking, hunting, washing and gathering wool, horse husbandry, fetching peas, laundry, and making clothes. One son had worked in service, and another at sea, and both were still contributing to the family economy; one daughter lived at home and baked bread, and another lived in Exeter, possibly in service, and appears to have either washed or mended clothes for her sister. What is more, none of these activities were undertaken by the male head of household, Thomas Conant, who was not examined and was not mentioned doing any work tasks in the depositions. Yet he is the only worker in the case who is given an occupational descriptor: husbandman. This reinforces the importance of looking beyond occupational titles when seeking to uncover a fuller picture of who did what in the preindustrial economy. Instead, we need to recognise a diverse cast of workers that centres married women, children and young people, servants, and others, as much as male heads of household with occupational titles.

As the case of the Conants highlights, both the work tasks undertaken and the types of people doing them were wide ranging in our period. This chapter aims to go beyond this important but rather general point to establish in more detail clear patterns of what types of people tended to engage in what types of tasks. To do so, it draws on the biographical details that witnesses provided when reporting work tasks.

2.1 Biographical Preambles

‘The Information of Joan Frend the wife of William Frend of Brixham labourer taken before me Ambrose Bellott Esquire … the 4th day of February 1619.’Footnote ²

Witness testimony in the early modern English courts was typically introduced by a preamble of this sort, providing details of when and by whom the deposition was taken, alongside biographical information about the deponent. These personal details, especially marital and occupational statuses, and ages when given, are of immense value to the work-task approach to the history of work, allowing the connections between certain types of work and the types of people who performed them to be identified. From her deposition we know that Joan Frend roasted some pilchards for a man who called at her alehouse, before going to a neighbour’s house to buy faggots of wood (during which time the man stole away with a shirt of her husband). Not only is there evidence here of women’s work tasks, these can also be classified as performed by a married woman, and by the wife of a labourer. Recording such information across the dataset allows for comparisons of the types of tasks done by women and men of different marital and occupational statuses, as well as examining the relationship between the work done by women and the occupation of their husband. It is also possible to compare the tasks performed by different age groups: although not given in this example from the quarter sessions, where age was rarely given, it was regularly included in church court depositions. In fact, none of this biographical information was consistently provided. As well as variations between courts, there were also variations over time and geography, with some clerks making a habit of recording certain things and not others. Most significantly, there were key gendered differences in what was recorded: occupational status was very rarely recorded for women; marital status was very rarely recorded for men. Nonetheless, we have been able to gather enough information about age, marital status, and occupation, to analyse the impact of all three on the experience of work.

It is important to do so, for historians of gender have long emphasised that exploring differences between women and men is insufficient for understanding the range of experiences of either. As Alexandra Shepard has argued, we need an understanding of gender ‘not solely premised on … dynamics between the sexes but which also takes into account the relations between different categories of men and of women’.Footnote ³ As her study of masculinity in early modern England demonstrated, patriarchal values privileged certain men over and above others, with middle-aged householders benefitting more than young men and poorer men. These differences were related to marital status, which was also of fundamental importance to shaping women’s life experiences. In her work on single women in early modern England, Amy Froide argued that marital status was ‘just as crucial’ as gender to understanding the operation of patriarchy, and the Gender and Work project has recently described marriage as of ‘immense importance’ to working life in early modern Sweden.Footnote ⁴ To understand the experience of work in early modern England, attention should be paid to these intersecting hierarchies that worked with gender to shape the lives of women and men.

Using biographical preambles to explore these hierarchies is not without its pitfalls. The accuracy of the ages provided by deponents must be questioned, as ages ending in nought are consistently overrepresented in reported age listings in the period. People reached for a convenient round number rather than absolute precision.Footnote ⁵ But as a rough guide to the relationship between age and work, they are still useful. The main limitation of marital status information is that it was not provided for men in preambles. In some instances, it is possible to identify married men if the testimony itself mentions a wife, but it is uncommon to find even these incidental references to the state of bachelor- or widower-hood, so it is not possible to compare the work done by men of differing marital statuses. Women’s marital status was commonly recorded across our sources in the form of ‘wife of’, ‘widow’, and for never-married women of any age, ‘spinster’ (though variants, such as ‘singlewoman’, were occasionally used instead). Whilst wife and widow are relatively straightforward designations, spinster does present some challenges of interpretation, because it also has occupational connotations of working as a spinner that complicate treating it simply as a marital status. Amy Erickson and Amy Froide have suggested that the use of spinster as an occupational descriptor was a medieval inheritance that had largely given way to its use as a marital status by the early modern period, but with the caveat that it did persist in some regions such as the north-west, even into the nineteenth century.Footnote ⁶ Here it is treated as a marital status, in line with its most common usage, except for instances where it appears alongside other contradictory evidence of marital status, in either a preamble or incidentally in a deposition, that suggests it was being deployed as an occupational status.Footnote ⁷

If the terms used to denote marital status contain elements of ambiguity, the same is even more true of occupational descriptors. They were commonly recorded for men, but precisely what they record is far from straightforward.Footnote ⁸ Many were hybrid designations of social and occupational status, such as yeoman or husbandman, whose precise relation to wealth could vary regionally and chronologically, and were rather vague as a guide to the specific work that individual might do. The same is true of terms like servant or labourer, which tell us more about employment relations than types of work. If some terms are unhelpfully imprecise, others convey a false sense of specialisation, with a title such as carpenter or shoemaker potentially failing to reflect ‘the piecemeal quality of many people’s existence’, which often drew on varied sources of income.Footnote ⁹ The descriptors that deponents laid claim to were just that: claims. They were built upon the realities of working life, but they were also intimately tied to individual and social identity formation, which renders them capricious guides to the working lives of their claimants.Footnote ¹⁰ There was, for instance, a tendency for individuals in more subordinate positions to ‘claim up’; servants often claimed a title such as husbandman in their preamble, only for further questioning to reveal their junior status, and a similar pattern can be found in the depositions of artisan apprentices, and labourers.Footnote ¹¹ Older men often still claimed a specialist occupational title, such as blacksmith, despite having ceased to make their living that way.Footnote ¹² In court cases where these tensions emerge, we are often left with multiple occupational descriptors for a single individual: described, for instance, as a husbandman in their preamble but in the course of their own deposition as ‘a day labourer’. These multiples not only complicate the logistics of categorisation and analysis; they provide a valuable reminder that certain occupational groups, such as servants and labourers, are likely to be underrepresented in our data. For every case where we can see a servant ‘claiming up’, there are likely to be many more where this claim was successfully enshrined in the legal record as fact. For this reason, when categorising all those who were identified as both a husbandman and a servant were grouped as the latter, to help balance out this effect, and on the basis that most of these individuals were in all likelihood servants in husbandry. Those identified as both husbandman and labourer have been classified differently, as multiples, because the precise relationship between these two claims is harder to establish, as is explored below.

Historians have become increasingly aware of the issues surrounding occupational descriptors and their complex relationship with the working lives of the individuals associated with them. But they are not meaningless, and the fact that they were so widely used by contemporaries makes it imperative to develop a fuller appreciation of their meaning. The work-task approach can play a key role here. By comparing occupational descriptors with the tasks done by the people claiming them, we can deepen our understanding of what these designations signified.

2.2 Working ‘For Another’

As well as seeking to establish the work tasks undertaken by different groups of people in early modern England, this chapter investigates the terms on which such work was undertaken. Interpretations of differences between key groups of working people – such as between husbandmen and labourers, or between married and unmarried women – have often hinged on the degree to which they were either required or able to undertake work that generated an income. Were husbandmen distinguished from labourers by their freedom from wage dependency and the need to sell their labour?Footnote ¹³ Did married women have greater or fewer opportunities to generate income for their household than unmarried women?Footnote ¹⁴

The work-task data does not record the exact terms under which all tasks were performed. Only 5 per cent of all tasks in the database include explicit references to individuals being paid for their labour or doing work that was formally contracted, a number that increases to 12 per cent when tasks explicitly undertaken as part of service are included. If most of the material is imprecise about the terms of employment under which work was done, it is not silent on the issue. What we have been able to record are frequent references to work undertaken ‘for another’ outside of the worker’s household unit. For instance, in 1631 John Smith of Bathwick, Somerset, deposed in a tithe case that he ‘at the request of the said William Fisher was the man which weighed his wool which were shorn from his sheep’.Footnote ¹⁵ One year later in the parish of Camerton in the same county, haymaking was done by Edward Bigg, who ‘came to the close and with his wife made the same up into bigger cocks for and in the name of Robert Eaton of Wellow’.Footnote ¹⁶ In thousands of other instances deponents reported similar examples of work done ‘for’ others, or which they had been ‘willed’ or ‘requested’ to do by other individuals, or offered other details, such as ‘reaping the corn of Widow Love’, that made it clear this was work done for an individual outside of their own household and not as direct subsistence work.

The language used by deponents to describe working for others was generally ambiguous about the precise employment relations involved. It is likely that in many cases such work was paid. In others it may have formed part of a reciprocal exchange of labour or goods. In some cases, it may have been an act of charity or kindness. But it is not possible to know the proportions of each. In many instances deponents made no reference to any form of employment relationship, simply reporting a task like ‘milking a cow’ with no further information, but that does not mean that such work was not undertaken ‘for another’, for pay or otherwise. Likewise, if an individual is recorded as making goods such as clothing or ale that may well have been sold, or a tailor is detailed repairing a pair of trousers, this could not be recorded as evidence of work done ‘for another’ unless the deponent explicitly stated they were making such goods or performing such services for a specific person outside of their household. Given the blurred lines between production for the household and production for the market, assumptions cannot be made about whether tasks such as these were income generating or not. As a result, it is likely that many more tasks were performed ‘for another’ than the proportion that have been explicitly identified. At the same time, it was rarely possible to be certain that any tasks were strictly subsistence only with no income-generating potential, and this was not therefore deployed as a category in our analysis.

The category of work undertaken ‘for another’ on unspecified terms can therefore only be taken as a very rough proxy of work that might have been income generating. But set alongside tasks that are known to have been explicitly paid, and work done as part of service, it provides another measure for thinking about work undertaken for others outside the household unit. In particular, it can be illuminating to compare the extent and variety of ‘for another’ work carried out by different groups of workers along lines of age, marital status, and occupation. To lay the foundation for such comparisons, Table 2.1 provides an overview of this data for the whole dataset. It shows the number and proportion of tasks in three ‘for another’ categories – explicitly paid, service, and unspecified terms – alongside the number and proportion of tasks for which there was no evidence of work being undertaken for an individual outside of the household. Taken together, the three ‘for another’ categories account for over one-third of all work tasks in the dataset, suggesting a relatively high proportion of work tasks that were conducted were potentially undertaken for payment of some sort. Significantly, although there is some difference in the distribution across the three categories for men’s and women’s ‘for another’ work, there is very little difference in the overall percentage of this work done by men, at 34.8 per cent, and by women, at 36.3 per cent, in our data. Nor do these patterns change over time: the overall percentage of ‘for another’ work for the period before 1630 is 36 per cent, and for the period after 1630 it is 35 per cent. If income-generating work was increasingly prevalent in this period, it is not reflected in this analysis.

The prevalence of ‘for another’ work across different areas of the economy is presented in Table 2.2, where the three ‘for another’ categories have been combined. For agriculture and land, or crafts and construction, the results are not especially surprising; these were types of work that were often done for others, but also often done as forms of work within the household unit as well. The most eye-catching figures here relate to carework, which was overwhelmingly ‘for another’ for both men and women. Carework, alongside housework, is the category of work most commonly characterised as unpaid ‘domestic’ work, existing outside of the market and the economy as they are conventionally conceived by economists. Our data calls for a major reappraisal of this assumption: care for the sick, the dying, newborns, and children was routinely something that individuals who were not part of that household were called on to perform, and it is likely they were often paid to do so. Some of this work was undertaken by servants, but this alone does not explain the high levels of ‘for another’ work undertaken in the category of carework. It is an issue returned to in Chapter 5, but it is worth noting here that carework and indeed housework were commonly performed ‘for another’.

Table 2.2Work tasks performed ‘for another’ by work category

Table 2.2Long description

The table consists of various labour work taken up by men and women. The first column represents the work types, which is followed by six columns on F tasks for another, percent of F tasks for another, M tasks for another, percent of M tasks for another, total tasks for another, and percent of total tasks for another. The data can be arranged as follows:

• For agriculture and land, the corresponding values are 189, 40, 1012, 46.8, 1201, and 45.6.

• For care work, the corresponding values are 279, 75.8, 162, 82.7, 441, and 78.2.

• For commerce, the corresponding values are 37, 6.7, 84, 5.4, 121, and 5.7.

• For crafts and construction, the corresponding values are 79, 38.3, 425, 57.7, 504, and 53.5.

• For food processing, the corresponding values are 42, 32.1, 147, 30.9, 189, and 31.1.

• For housework, the corresponding values are 251, 48, 76, 33.8, 327, and 43.7.

• For management, the corresponding values are 17, 9.9, 67, 14.2, 84 and 13.

• For transport, the corresponding values are 74, 35.6, 400, 38.9, 474, and 38.3.

• For other, the corresponding values are 7, 14.3, 48, 44.4, 55, and 35.

• The total tasks in all were 975, 36.3, 2421, 34.8, 3396, and 35.2.

This chapter now turns to the main categories of analysis that biographical preambles permit us to explore when combined with the work-task data. First, it analyses the ways in which repertoires of work tasks and patterns of ‘for another’ work changed across the lifecycle and were affected by marital status. Second, it explores the relationship between occupations and work tasks across two sections: one comparing the relative sectoral distributions of each, engaging with the PST (primary, secondary, tertiary) classification system, and one exploring at length the relationship between different occupational groups and their work repertoires.

2.3 Age and Marital Status

In 1653, John Saunders of Ottery St Mary, in East Devon, gave to his daughter Anne, who lived with him, a goose and a gander, for ‘the keeping and increase of the said geese’.Footnote ¹⁷ Anne was just three or four years old at the time but was already being asked to enter the world of work and take responsibility for the care of a pair of geese. Anne was not the only child of this age put to work. In July of 1629 the four-year-old Thomas Britten of Upton Noble, Somerset, was sent by his mother as a ‘messenger’ to fetch her sister-in-law to come to their house to buy a parcel of yarn.Footnote ¹⁸ If Elizabeth Twise of Northenden, Cheshire, started so young, her working life would have begun in the reign of Queen Elizabeth and still been going strong well into that of Charles II: at the age of ‘80 and upwards’ she was delivering babies as a midwife in 1674.Footnote ¹⁹ Work shaped the lives of early modern women and men from some of their earliest years to their last. When Elizabeth Upham died in Crowcombe, Somerset, in 1620, her neighbour John Welshowe confirmed that ‘the said Elizabeth was near 80 years old before her death and did spin and card hard until shortly before her death’.Footnote ²⁰

These examples represent some of the youngest and oldest workers recorded in the database, but as Table 2.3 shows, the work-task data’s main strength is that it captures a good spread across all the ages in between. Ages are not always provided, but we are able to identify the age of the actor for precisely 20 per cent of all work tasks, just under 2,000 tasks, with similar proportions for women and men, and an average age of 36 for both.Footnote ²¹

The smallest number of tasks recorded were those undertaken by the very old and the very young. That is a reflection of demographic structure in the case of the very old: they were a small proportion of people. Wrigley and Schofield estimated that the proportion of the population over 60 in the seventeenth century was between 8 and 9 per cent; the number of tasks in the data performed by people aged 60 and over is 9 per cent. The youngest age group is undoubtedly underrepresented: Wrigley and Schofield estimated they made up between 31 and 34 per cent of the population, but they performed just 6 per cent of tasks in our data. This may well reflect their reduced economic contribution relative to other groups but is also a consequence of young children being far less likely to appear as witnesses before the courts and to detail their own work tasks.Footnote ²² Our 15 to 24 category is in line with the demographic data that puts it at 17–18 per cent of the population, but our ‘middle age’ groupings are overrepresented. Wrigley and Schofield estimate 25 to 59-year-olds as 40–43 per cent of the population, whereas they account for two-thirds of all of our work tasks.Footnote ²³ To some extent this reflects their high levels of economic activity compared to the very young and the old, but they were also more likely to depose than young adults who carried less status.

Whilst the category of children’s work is a relatively small one, it is worth exploring the material that is available, given how difficult this subject is to study. One clear pattern is that the work of both young boys and young girls was much more heavily concentrated in a small number of categories than the work repertoires of adults. For boys under the age of 15, the point at which they are likely to have entered service, agricultural work dominated, and within that animal husbandry in particular. Boys fed oxen, herded sheep, and drove geese and pigs. They were also employed to do a range of fetching and carrying jobs. In addition to taking messages, they collected firewood, drew water from wells, and might in their early teens be entrusted with driving carts, especially within the farm. They were also commonly charged with guarding crops against birds, which included climbing trees to destroy the nests of crows and magpies. The work of young girls was concentrated within the category of housework. They made beds, washed food and tableware, and above all collected water, a task which would often have taken them outside of the household, and was a physically demanding one, so it is perhaps surprising to see this subcategory dominating the work of young girls. They also engaged in some agricultural tasks: keeping geese, milking cows, and weeding fields. Amanda Flather has suggested that ‘little girls and boys parted company very early’ in their working lives, with their work concentrated in different spaces.Footnote ²⁴ Whilst there may have been opportunities for encounters between them at the local well or river when collecting water, this conclusion generally holds true. In fact, with over 50 per cent of boys’ work concentrated in agriculture, and closer to 60 per cent of girls’ work in housework, the differences between male and female work repertoires were more pronounced for children than they were for adults. As Flather puts it, ‘gender boundaries could be blurred by age’; they generally started out as quite rigid for the youngest of workers.Footnote ²⁵

Table 2.4 shows the average work repertoires for women at different stages of the lifecycle, demonstrating that women’s work became more diverse as they reached adulthood. The age groupings used here are broader than those used in Table 2.3 to ensure that a reasonable number of tasks can be attributed to each, but they are also intended to capture key phases of the lifecycle: 0 to 24 was a period of childhood followed for most by a period of service before marriage; 25 to 44 would have represented the main period of childbearing and raising young children for most women; and 45 and over a stage of life where the duty of care for young children would have eased as those children grew more independent.

Table 2.4Women’s lifecycle repertoires

Table 2.4Long description

The table indicates various age groups and their corresponding statistical values. The first column lists various labour work, and the subsequent three columns are age groups of 0 to 24, 25 to 44, and 45 plus. They are further sub divided into tasks, and repertoire in percent. The data is arranged as follows:

1. 1. 0 to 24 age group task distribution:

   • For agriculture and land, the corresponding values are 34 and 24.6.

   • For carework, the corresponding values are 11 and 8.

   • For commerce, the corresponding values are 8 and 5.8.

   • For crafts and construction, the corresponding values are 16 and 11.6.

   • For food processing, the corresponding values are 2 and 1.4.

   • For housework, the corresponding values are 54 and 39.1.

   • For management, the corresponding values are 1 and 0.7.

   • For transport, the corresponding values are 9 and 6.5.

   • For other, the corresponding values are 3 and 2.2.

   • For total, the corresponding values are 138 and 99.9.

2. 2. 25 to 44 age group task distribution:

   • For agriculture and land, the corresponding values are 39 and 20.2.

   • For carework, the corresponding values are 41 and 21.2.

   • For commerce, the corresponding values are 25 and 13.

   • For crafts and construction, the corresponding values are 25 and 13.

   • For food processing, the corresponding values are 3 and 1.6.

   • For housework, the corresponding values are 28 and 14.5.

   • For management, the corresponding values are 10 and 5.2.

   • For transport, the corresponding values are 15 and 7.8.

   • For other, the corresponding values are 7 and 3.6.

   • For total, the corresponding values are 193 and 100.1.

3. 3. 45 plus age group task distribution:

   • For agriculture and land, the corresponding values are 31 and 20.5.

   • For carework, the corresponding values are 57 and 37.7.

   • For commerce, the corresponding values are 9 and 6.

   • For crafts and construction, the corresponding values are 8 and 5.3.

   • For food processing, the corresponding values are 1 and 0.7.

   • For housework, the corresponding values are 27 and 17.9.

   • For management, the corresponding values are 8 and 5.3.

   • For transport, the corresponding values are 8 and 5.3.

   • For other, the corresponding values are 2 and 1.3.

   • For total, the corresponding values are 151 and 100.

Agricultural activities formed a significant part of women’s work repertoires throughout the lifecycle, but beyond that the balance of tasks women undertook shifted considerably as they aged. The most striking pattern here is the significant rise in carework across the life course, which became prominent in the main child-raising phase of life but continued, and indeed increased, in importance in later life. The proportion of housework trended in the opposite direction; it was the largest category of work for children and for service-age women but formed a less substantial component of the work repertoires of those 25 and above. These changes are further illuminated by looking at the main subcategories of work at different stages of the lifecycle, as in Table 2.5.

For younger women the category of housework that loomed large in their repertoires was itself dominated by the collecting of water for household duties, and the provision of food and drink, which included both preparing and serving food. Their agricultural work was focused on animal husbandry. For women in ‘middle age’ the character of agricultural work shifted, with fieldwork becoming more prominent, suggesting that the opportunity to undertake seasonal summer work was an important one for married mothers.Footnote ²⁶ As we might expect, childcare became an important part of the work repertoire at this life-stage, but the upsurge in carework was also due to a rise in the provision of healthcare. This was a subcategory that remained important for older women, alongside that of midwifery, both forms of work that required the experience and skill that came with age.Footnote ²⁷ Older women’s agricultural activity represented a blend of animal husbandry and fieldwork, bringing together tasks that had dominated at different stages of the lifecycle. The care of animals, to some extent, took a back seat in mid-life when childcare made greater demands, though wives still did a higher proportion of all animal husbandry than either servants or widows because there were more of them.Footnote ²⁸ Textile production, and especially spinning, featured in women’s work repertoires across their lives. Although it was a smaller subcategory for older women (7 tasks), it was still a notable one, and the database includes examples of women between the ages of 58 and 80 scouring, spinning, and carding wool.

Age was an important factor in shaping the work tasks that women undertook in early modern England. The same is less true for men. Table 2.6 demonstrates that agriculture remained the largest category of men’s work throughout their lifecycle. Involvement in commerce and management was more common for men over the age of 25, whereas more time-consuming and menial transportation work played a larger part in the working lives of younger men.

Table 2.6Men’s lifecycle repertoires

Table 2.6Long description

The table indicates various age groups and their corresponding statistical values. The first column lists various labour work, and the subsequent three columns are age groups of 0 to 24, 25 to 44, and 45 plus. They are further sub divided into tasks, and repertoire in percent. The data is arranged as follows:

1. 1. 0 to 24 age group task distribution:

   • For agriculture and land, the corresponding values are 169 and 49.6.

   • For carework, the corresponding values are 14 and 4.1.

   • For commerce, the corresponding values are 17 and 5.

   • For crafts and construction, the corresponding values are 34 and 10.

   • For food processing, the corresponding values are 17 and 5.

   • For housework, the corresponding values are 20 and 5.9.

   • For management, the corresponding values are 6 and 1.8.

   • For transport, the corresponding values are 60 and 17.6.

   • For other, the corresponding values are 4 and 1.2.

   • For total, the corresponding values are 341 and 100.2.

2. 2. 25 to 44 age group task distribution:

   • For agriculture and land, the corresponding values are 333 and 48.4.

   • For carework, the corresponding values are 28 and 4.1.

   • For commerce, the corresponding values are 102 and 14.8.

   • For crafts and construction, the corresponding values are 92 and 13.4.

   • For food processing, the corresponding values are 11 and 1.6.

   • For housework, the corresponding values are 11 and 1.6.

   • For management, the corresponding values are 46 and 6.7.

   • For transport, the corresponding values are 60 and 8.7.

   • For other, the corresponding values are 5 and 0.7.

   • For total, the corresponding values are 688 and 100.

3. 3. 45 plus age group task distribution:

   • For agriculture and land, the corresponding values are 166 and 39.8.

   • For carework, the corresponding values are 19 and 4.6.

   • For commerce, the corresponding values are 59 and 14.1.

   • For crafts and construction, the corresponding values are 69 and 16.5.

   • For food processing, the corresponding values are 20 and 4.8.

   • For housework, the corresponding values are 5 and 1.2.

   • For management, the corresponding values are 29 and 7.

   • For transport, the corresponding values are 43 and 10.3.

   • For other, the corresponding values are 7 and 1.7.

   • For total, the corresponding values are 417 and 100.

At the level of subcategories, continuity across the lifecycle outweighs change. Fieldwork dominated for all ages of men, with animal husbandry and farm transport figuring in the top four subcategories for each group too. Younger men were more likely to be involved in transport work, and especially that which involved driving carts, whereas the two older groups of men were instead more likely to be involved in building work, which included thatching, masonry, and assessing buildings for repairs. As with women’s work, there was a shift across the lifecycle towards work that required more authority (management), experience (commerce), or skill (buildings), and away from certain forms of more menial physical work (carting), but these changes were muted, with agricultural activity constantly outnumbering other types of tasks for all men. Age did influence the types of tasks men undertook but in ways that were less pronounced than for women.

Did patterns of work ‘for another’ change across the lifecycle? As Table 2.7 shows, the rates of such work were lower for children than for all other age groups, as their work was more likely to be performed at the request of another member of the household unit, especially parents. The percentages are nonetheless high, with around a third of children’s work done ‘for another’, but that is in large part accounted for by a number of ‘service’ activities in the database performed by children under 14. Beyond that we can identify a pattern, for both women and men, of a greater prevalence of working ‘for another’ during the main ages of service, and again after the main phase of childrearing had passed, than during the ‘middle’ phase of life, though this still was an important period of performing work for people outside of the household. Those in the oldest age group performed less work ‘for another’, perhaps because they were less able to.

These patterns can be interrogated further by considering the main categories of ‘for another’ work undertaken by women and men. The main source of ‘for another’ work for women changed across the life cycle, from housework when young/service age (0–24), to agriculture in middle age (25–44), to carework when older (45+). For women in middle age, the overwhelming majority of agricultural tasks performed were undertaken ‘for another’, at 85 per cent, again suggesting that for women in this age bracket working in agriculture, especially in seasonal fieldwork, served as an important source of income.Footnote ²⁹ Moreover, the figures warn against assuming that women in this age group had reduced opportunities to undertake income-generating work because of the burden of carework that fell on them in these years. Instead, much of the carework they performed, especially healthcare, was done ‘for another’. It may be that many married women in this grouping were able to monetise skills they developed providing care for their own families to bring in additional income by offering care services to others. They continued to perform this work ‘for another’ in later life too. Turning to men’s work ‘for another’, such changes are less pronounced: agriculture was the main source of such work for men of all ages. Age and experience had less of an impact on what men did ‘for another’. It may have influenced what they were paid or on what terms, but the work-task data does not reveal this information.

Was marital status ‘as crucial as gender’ in shaping the experience of work in early modern England? There is a close correlation between marital status and different lifecycle stages, but not all unmarried people were young, and not all widows and widowers were old people, and vice versa, so it is important to analyse the relationship between work and marital status separately from age. It is not possible to do this for the men in our database. Marital status can only be identified for around 10 per cent of men, and for over 90 per cent of these that status is married, as it has been deduced from reference to a wife in their deposition. For tasks performed by women, on the other hand, marital status is noted in two-thirds of instances, providing a healthy sample of just under 1,800 tasks for analysis. Of those, 21 per cent were performed by never-married women, 64 per cent by married women, and 15 per cent by widows. This underrepresents the work of never-married women, who accounted for somewhere between 30 and 35 per cent of the adult female population in this period, and overrepresents that of married women, who accounted for closer to 50 per cent, whilst being roughly in proportion for widows.Footnote ³⁰ Nonetheless, it does allow meaningful comparisons of the work repertoires of women of each marital status, as shown in Table 2.8.

Table 2.8Women’s repertoires by marital status

Table 2.8Long description

The table depicts the marital status of women. The first column represents various labour work, followed by six columns with data for never married tasks, never married repertoire in percent, married tasks, married repertoire in percent, widowed tasks, and widowed repertoire in percent. The data is as follows:

• For agriculture and land, the corresponding values are 85, 22.8, 175, 15.3, 36, and 13.3.

• For care work, the corresponding values are 30, 8.0, 145, 12.7, 62, and 23.

• For commerce, the corresponding values are 66, 17.7, 275, 24.1, 68, and 25.2.

• For crafts and construction, the corresponding values are 29, 7.8, 87, 7.6, 20, and 7.4.

• For food processing, the corresponding values are 14, 3.8, 62, 5.4, 17, and 6.3.

• For housework, the corresponding values are 88, 23.6, 209, 18.3, 29, and 10.7.

• For management, the corresponding values are 14, 3.8, 98, 8.6, 15, and 5.6.

• For transport, the corresponding values are 43, 11.5, 79, 6.9, 17, and 6.3.

• For other, the corresponding values are 4, 1.1, 11, 1, 6, and 2.2.

• For total, the corresponding values are 373, 100.1, 1141, 99.9, 270, and 100.

This data shows a clearer distinction between the proportions of agricultural work undertaken by different groups of women than in the age group analysis, implying that it was marriage rather than age per se that reduced the proportion of agricultural work women undertook. Other patterns here accord more closely with the age data, with a clear rise in the proportion of carework aligning with changes in marital status in ways that mirror lifecycle developments. Women who were either married or widowed were more heavily involved in commercial and management activities, and dedicated less of their working lives to housework and transport tasks, than single women. Much of this is in line with the findings for early modern Sweden, where Ågren et al. have emphasised the importance of marriage for creating hierarchies between women’s experiences of work.Footnote ³¹ In England, never-married women were more likely to engage in menial, arduous, and time-consuming tasks than other women: they washed sheep, checked on cattle, took clothes to be mended, took corn to the mill, took linen to be washed, cleaned rooms, collected water, served drinks, washed clothes, and milked cows; very often as servants in subordinate roles to married women. Married women might also perform any of these tasks, but they did so less frequently, and they also took on work that required more skill or authority: they sheared sheep; they butchered sheep; they purchased, sold, and pawned goods; they delivered and nursed babies; made clothes; settled debts and lent money; brewed ale and beer; and treated the sick and dying. They were also more likely to direct the work of others, both artisans and their servants, such as arranging for keys to be made, or pies to be baked.

The tasks performed by widows were closer to those done by married women than they were to those by never-married women, supporting the argument of Ågren et al. that it is the distinction between never-married and ever-married women that mattered most. If the work tasks of widows appear similar to those of married women, the picture of their work repertoires presented here does differ in interesting ways from that obtained by other approaches to widows’ work. Work on probate inventories highlights caring for livestock, food processing, textile production, and money lending as key work categories for enterprising widows.Footnote ³² The database does include examples of widows droving sheep and treating sick colts, brewing beer and winnowing barley, spinning yarn and carrying wool, and lending money and making and receiving payments, but none of these represent major categories of widows’ work. Instead, the depositional material captures aspects of widows’ work that would have been less likely to leave a material trace in the probate record: their buying and selling of goods and their extensive involvement in midwifery. This material adds a valuable dimension to our understanding of how widows were able to run households, with commerce and carework very prominent in their work repertoires.

The conclusion that never-married women were more likely to do work that was lower-skilled, more menial, and lacking in authority relative to the work of ever-married women is not surprising, given that most of these single women were young and working in service. This is a reasonable assumption, but it is important to recognise that not all unmarried women would have fallen into this category. Amy Froide’s work has been particularly important for exploring the working lives of those women who remained single into adulthood, especially in urban contexts.Footnote ³³ The database contains 373 tasks performed by never-married women, but we can only identify 32 of these as being undertaken by never-married women over the age of 25, half of which were performed by women in their late 20s, many of whom were still working in service. Of the 16 done by never-married women aged 30 or over, five were also explicitly done as part of work in service, a useful reminder that not all servants were young.Footnote ³⁴ One of the remaining examples is Catherine Edwards, a 35-year-old spinster living in Crediton, Devon, in 1674, ‘being employed there for some few weeks as nursekeeper to Zachary Smith’s wife then in childbed’.Footnote ³⁵ Another was Elizabeth Luenden, 30 years of age and working in Gosport, Hampshire, in 1590, ‘attending in her sickness’ one Elizabeth Spenser whom she had used to work alongside as a fellow servant.Footnote ³⁶ These examples suggest that never-married women who had left service might, like middle-aged and older women who had married, have been able to undertake carework for pay. But both instances come from relatively large towns, and may well be indicators, alongside the lack of other examples of the work of older never-married women in rural contexts, that options were sufficiently limited to push older single women to move into urban environments to find work.

Never-married women were, as Table 2.9 suggests, more likely to be engaged in work ‘for another’ than their ever-married counterparts: 45.0 per cent of their work tasks were undertaken for someone outside of their household or as a servant. There are clearer differences here between married women, for whom the figure was just 25.2 per cent, and widows, for whom it was 38.5 per cent. If they performed similar work repertoires, widows were more likely to do these tasks ‘for another’. This has interesting implications for how married women’s work is characterised. Whilst the Gender and Work project has argued that ‘marriage facilitated rather than prevented women’s work and income-earning’, in the English context married women were less likely to undertake work for people outside of their household, reducing one form of income-generating work. It is important to bear in mind the limitations of our ‘for another’ category here though, as it does not reliably capture all forms of income generation, such as producing and selling goods on the market, so we cannot conclude that married women generated less income through their work than women outside of marriage.Footnote ³⁷ These differences in work ‘for another’ could therefore be interpreted positively, as evidence married women were more fully engaged in income-generating work within the household, or negatively as a consequence of coverture.

Table 2.9Women’s work performed ‘for another’ by marital status

Table 2.9Long description

The table consists of various labour work taken up by marital women for others. The first column represents the work types, which is followed by six columns on never married tasks for another, percent of never married tasks for another, married tasks for another, percent of married tasks for another, widowed tasks for another, percent of widowed tasks for another. The data can be arranged as follows:

• For agriculture and land, the corresponding values are 36, 42.4, 51, 29.1, 14, 38.9.

• For carework, the corresponding values are 21, 70, 105, 72.4, 48, 77.4.

• For commerce, the corresponding values are 10, 15.2, 8, 2.9, 3, 4.4.

• For crafts and construction, the corresponding values are 19, 65.5, 17, 19.5, 7, 35.

• For food processing, the corresponding values are 4, 28.6, 9, 14.5, 4, 23.5.

• For housework, the corresponding values are 56, 63.6, 68, 32.5, 19, 65.5.

• For management, the corresponding values are 4, 28.6, 7, 7.1, 1, 6.7.

• For transport, the corresponding values are 18, 41.9, 18, 22.8, 8, 47.1.

• For other, the corresponding values are 0, 0, 4, 36.4, 0, 0.

• For total, the corresponding values are 168, 45.0, 287, 25.2, 104, 38.5.

For women, marital status was a key factor in shaping the experience of work in early modern England, in ways that were closely related to age and the lifecycle. For the young and the unmarried work was lower status and lower skilled, and often performed ‘for another’. Marriage and ‘middle age’ brought women more authority, and tended to involve more skilled work activities, alongside a partial retreat from working for others. Older age or widowhood did not entail a significant shift in the repertoire of tasks done, but the terms on which they were performed changed once again, with a return to doing more work for others, although much of this was now carework that utilised skills and experience obtained in mid-life. The data reveals little about the impact of marital status on men’s working lives, but if it too follows similar patterns to the lifecycle analysis then it was far more muted than for women’s. There may have been similar changes in the amount of work done ‘for another’ according to marital status, but the repertoires of work that men performed across their lives were relatively stable.

2.4 Occupations, Tasks, and Sectors

Occupational descriptors have played a fundamental role in the history of work. Their use by contemporaries across a range of legal and administrative records, as well as in cultural mediums such as plays and cheap print, make them the most ubiquitous indicators of the work that was undertaken by people in the past. Their abundance in the historical record has made them particularly valuable to quantitative approaches to historical work patterns, most notably serving as the focal point in recent discussions of structural economic change in early modern England.Footnote ³⁸ As discussed above, however, there are questions about the reliability of occupational descriptors as guides to the work individuals actually did, and therefore about the value of quantifying them to measure economic change. The work-task approach provides a way to bridge these two perspectives, allowing the relationship between occupational descriptors and the work done by their claimants to be investigated, whilst still attempting meaningful quantification of patterns of work across society.

One limitation of occupational descriptors that is difficult to overcome is that they were rarely provided for women. There are two exceptions: servants and midwives. Of the just under 2,000 female actors (those who performed work tasks) recorded in the database there are occupational descriptors for just 341, or 17 per cent. Of those, 62 per cent were servants and 28 per cent were midwives. The remaining 10 per cent included baker, seamstress, alehouse-keeper, chandler, peddler, bookseller, cook, hosier, housekeeper, and stockinger. As a consequence, the data relating to female occupations is very uneven and heavily concentrated on just two occupations, and so not especially useful for assessing differences in women’s work. Nonetheless, some analysis is attempted later in this chapter, but in the first instance the focus is on male occupations, which provide richer data: 56 per cent of male actors have an occupational or social descriptor, just over 2,700 individuals. These have been grouped into occupational categories in Table 2.10.

Table 2.10Social/occupational status of male actors in the dataset

Notes: Most of these categories are self-explanatory and directly reflect the descriptors given, but others are composites of less common occupations. Professions include clerical, medical, legal, and educational titles; commercial trades include merchants, shopkeepers, victuallers, and other retailers; agricultural trades include shepherd, fisherman, gardener, shearman, ploughman, and other specified occupations directly involved in agricultural work; transport trades include carriers, carters, sailors, drovers, ostlers, and other occupations directly involved in transport; multiple includes instances where two descriptors were given for an individual from two different categories, and to which the rules specified in Section 2.1 were applied. The totals in this table are higher than those in Table 1.1, which only includes men who deposed before the courts, whereas this incorporates all men who were witnessed performing tasks.

Table 2.10Long description

The table depicts data for occupational status of men. The first column indicates the various social statuses, followed by two columns with data for number and percentage.

• For gentry, the corresponding values are 52 and 1.9.

• For professions, the corresponding values are 82 and 3.

• For yeomen, the corresponding values are 140 and 5.1.

• For commercial trades, the corresponding values are 75 and 2.8.

• For artisans, the corresponding values are 820 and 30.2.

• For husbandmen, the corresponding values are 536 and 19.7.

• For agricultural trades, the corresponding values are 76 and 2.8.

• For transport trades, the corresponding values are 43 and 1.6.

• For labourers, the corresponding values are 432 and 15.9.

• For servants, the corresponding values are 320 and 11.8.

• For apprentices, the corresponding values are 26 and 1.

• For multiple, the corresponding values are 46 and 1.7.

• For miscellaneous, the corresponding values are 71 and 2.6.

• For total, the corresponding values are 2719 and 100.

As was the case with male deponents discussed in Chapter 1 this is a fairly representative cross-section of the male population of early modern England. The possible exception is the underrepresentation of labourers. It has already been noted above that this may be due to a reluctance on the part of labouring men to claim such a descriptor if they could instead pass as a husbandman before the courts, but it may also be due to regional variations and the composition of our sample. Almost half of these actors are drawn from our south-west sample, where only 7 per cent of men with descriptors were styled as labourer. In the northern sample this figure was 20 per cent, and in the Eastern sample 28 per cent, which is in keeping with what historians know about regional patterns of proletarianisation.Footnote ³⁹ It is the dominance of the south-west material in our dataset, where dependence on wage labour was less common, that makes the percentage of labourers appear relatively low overall.

Another way to interrogate how representative the male workers in the work-task dataset are is to compare the occupational profile with that of the large sample of probate data assembled by Sebastiaan Keibek for his PhD as part of The Cambridge Group for The History of Population and Social Structure’s (CamPop) occupational structure project.Footnote ⁴⁰ Central to the approach of CamPop and others investigating structural change in the early modern economy is the grouping of occupational descriptors into the primary, secondary, and tertiary sectors to establish the proportion of men working in these different key areas of the economy.Footnote ⁴¹ The same can be done with our data to provide a comparison. Occupations were assigned to sectors using CamPop’s PST look-up table.Footnote ⁴² Assigning servants and labourers to sectors is more complicated as their allocation to a sector depends on what type of work they were employed to do. This is a problem Keibek dealt with in his PhD. To produce comparable results, we divided servants and labourers between sectors using the same proportions as Keibek. The servants of upper-status individuals are split 40:60 between farm servants (primary) and domestic servants (tertiary), with all other servants considered to be farm servants (primary); 10 per cent of labourers have been placed in the secondary sector, with the rest in the primary sector.Footnote ⁴³ The results of this classification are shown in Table 2.11.

Table 2.11Male actors’ occupations by sector

Notes: This excludes six male actors whose occupations were unclassifiable, hence the different total to Table 2.10. Keibek’s data is from ‘Male occupational structure’ p. 152. His figures for 1621 and 1641 have been averaged to provide an estimate for 1631.

Table 2.11Long description

The table depicts data for occupations by men in various sectors. The first column indicates the sector, followed by three columns with data for number of occupations, percent of occupations, and Keibek’s data, 1631 in percent.

• For primary sector, the corresponding values are 1,515, 55.8, and 59.

• For secondary sector, the corresponding values are 893, 32.9, and 32.4.

• For tertiary sector, the corresponding values are 305, 11.2, and 8.7.

• The total gives the corresponding values as 2713, 99.9, and 100.1.

As a point of comparison, Keibek’s sectoral distribution has been taken for the year 1631, which represents an approximate median for our dataset with roughly equal numbers of tasks from before and after this date. The results are reassuringly similar, with the secondary sector very closely aligned, and a slightly larger tertiary sector in our data compared to Keibek’s, but the overall proportions do not differ substantially. They are also similar to the proportions obtained by Keibek applying a PST analysis to the 1608 Gloucestershire Muster Roll, which results in a primary sector of between 54 and 58 per cent, a secondary sector of 35–37 per cent, and a tertiary sector of 6–8 per cent.Footnote ⁴⁴ The proportion of men in the work-task dataset with occupations in each sector, based on occupational title, is therefore very similar to measures of the occupational structure of the economy at this time deduced from completely different sources. But what does the sectoral distribution look like if the tasks performed are classified rather than occupational titles?

To classify tasks each of the 62 subcategories of work tasks were assigned to a sector, using the PST look-up table for occupational titles associated with work in that subcategory.Footnote ⁴⁵ At this stage it also makes sense to apply one of the tools discussed in the methodology chapter and filter the data by information status. Removing those tasks classed as integral, the tasks which we know to be most heavily shaped and likely overrepresented by certain types of court cases, provides a more accurate picture of the actual distribution of tasks across these subcategories in the economy. To produce the overall total of tasks a multiplier has been applied to the totals for women’s work tasks so that half of all work tasks were undertaken by women. The results are shown in Table 2.12.

Table 2.12Sectoral distribution of tasks

Notes: Integral excluded; adj. = adjusted (x2.45). The multiplier here differs from the standard x2.59 multiplier, as it is intended to create an equal number of male and female tasks remaining after integral tasks have been excluded from this analysis.

Table 2.12Long description

The table depicts data for tasks by men and women in various sectors. The first column indicates the sector, followed by six columns with data for tasks, percentage, F tasks adjusted in percent, percentage, M tasks, and percentage.

• For primary sector, the corresponding values are 3,077, 31.5, 1041, 21.3, 2036, and 41.6.

• For secondary sector, the corresponding values are 1,239, 12.7, 546, 11.2, 693, and 14.2.

• For tertiary sector, the corresponding values are 5,461, 55.9, 3300, 67.5, 2161, and 44.2.

• The total gives the corresponding values as 9778, 100.1, 4888, 100, 4890, and 100.

This presents a very different perspective on the sectoral distribution of work in the early modern English economy from an analysis based on occupational descriptors, with a dramatically larger tertiary sector comprised of high numbers of tasks performed in commerce and transport for both genders, and in carework and housework for women. In certain respects, this striking contrast is not difficult to explain: women’s tasks were more heavily concentrated in the tertiary sector, so introducing women’s work into the overall totals inflates that sector compared to the results based solely on male occupational descriptors presented in Table 2.11. But it is also clear that men’s occupations significantly underrepresent their work in the tertiary sector in comparison to work tasks. It is also likely that the work-task data captures more unpaid work than occupations do. Overall, work tasks offer a more complete view of the economy incorporating forms of work, such as women’s work and unpaid work, that have been overlooked in conventional approaches to economics and economic history. For comparison, we can analyse the sectoral distribution of tasks that were performed ‘for another’ and were therefore more likely to have been income-generating work, as shown in Table 2.13.

Table 2.13Sectoral distribution of tasks performed ‘for another’

Notes: Integral excluded; adj. = adjusted (x2.45).

Table 2.13Long description

The table depicts data for tasks by men and women in various sectors. The first column indicates the sector, followed by six columns with data for tasks for another, percentage, F tasks for another adjusted, percentage, M tasks for another, and percentage.

• For Primary sector, the corresponding values are 1479, 34.7, 453, 20.8, 1026, and 49.2.

• For Secondary sector, the corresponding values are 656, 15.4, 230, 10.6, 426, and 20.4.

• For Tertiary sector, the corresponding values are 2124, 49.9, 1492, 68.6, 632, and 30.3.

• For Total, the corresponding values are 4259, 100, 2175, 100, 2084, and 99.9.

This attempt to align our data more closely with forms of income-generating work does little to change the distribution of women’s work across sectors. This is an important finding in itself, highlighting that women’s tertiary work cannot simply be dismissed as ‘unpaid’. It does however serve to reduce the proportion of men’s tasks in the tertiary sector, mainly by removing most of the commercial tasks, especially buying and selling, which were seldom recorded as done ‘for another’. The sector nevertheless remains roughly three times the size indicated by occupational descriptors, largely due to the importance of various transport activities such as carting, carrying goods, and droving.Footnote ⁴⁶ Another adjustment that can be made is to remove the categories of carework and housework altogether. Studies of occupational structure largely exclude these types of work, assuming they are unpaid work for the maintenance of the worker’s own family, which falls outside the definition of labour force participation used in economics.Footnote ⁴⁷ We reject this characterisation of carework and housework, which often were paid in this period and certainly fall within the third-party criterion definition used in this book, but for the purposes of a closer comparison between the datasets it can be filtered out of the work-task data, as in Table 2.14.

Table 2.14Sectoral distribution of ‘for another’ tasks excluding carework and housework

Notes: Integral excluded; adj. = adjusted (x2.45).

Table 2.14Long description

The table depicts data for tasks excluding carework and housework by men and women in various sectors. The first column indicates the sector, followed by six columns with data for tasks for another excluding carework or housework, percentage, F tasks for another adjusted, excluding carework or housework, percentage, M tasks for another excluding carework or housework, and percentage.

• For Primary sector, the corresponding values are 1479, 52.7, 453, 48.6, 1026, 54.8.

• For Secondary sector, the corresponding values are 656, 23.4, 230, 24.7, 426, 22.8.

• For Tertiary sector, the corresponding values are 670, 23.9, 250, 26.8, 420, 22.4.

• For Total, the corresponding values are 2805, 100, 933, 100.1, 1872, 100.

The impact on the sectoral distribution of women’s work is dramatic, as these are very large categories within women’s work repertoires, and this in turn has an impact on the overall figures. With these additional adjustments, the primary sector for men, based on tasks, is now in proportion to the primary sector based on occupational titles alone. However, the tertiary sector remains considerably larger, and the secondary sector smaller, than we would expect from occupational data. This analysis shows that the work-task approach requires a revision of our understanding of the importance of the tertiary sector in this period, not only because it was prominent in women’s work but also a crucial part of men’s work. Moreover, a key reason why the secondary sector appears relatively small in the sectoral distribution from tasks is that many of the tasks performed by men with secondary sector occupational descriptors were actually tertiary tasks. It is to the relationship between descriptors and tasks that the following section now turns.

2.5 Occupations, Repertoires, and By-Employment

A valuable recent contribution to debates about the utility of occupational and social descriptors has been research that uses qualitative material to investigate what individuals with certain descriptors actually did in their working lives. This can be approached at the level of individual examples, or in some cases at scale when larger datasets are available.Footnote ⁴⁸ Research by the Gender and Work project, and by Alexandra Shepard, has shown that the relationship between occupations and ways of making a living was not straightforward, with descriptors often concealing the piecemeal character of working life. In these examples the focus has been on comparing descriptors with the full range of different ways of making a living or maintaining oneself and family that were deployed by men and women. Here, we add to this work by focusing on the relationship between descriptors and the specific tasks that people undertook.

Individual examples are suggestive here. The 21-year-old William Ricketts was a resident of East Knoyle, Wiltshire, in the year 1600. When giving a deposition in a local tithe dispute, he told the clerk that he was ‘a carpenter and doth that way wholly employ himself’. And yet, elsewhere in his statement, he recounted how ‘at harvest last’ he had been employed by a Mr Goldsborowe to ‘make up into hay’ a piece of ground, the which hay he ‘dried and made up into cocks’ before setting out the appropriate amount for tithe collection.Footnote ⁴⁹ Carpentry clearly was not his whole employment. On a Saturday morning in September of 1618, Owen Perfitt of West Cranmore, Somerset, ‘came into the field to fetch home some wheat’, where he likewise noted some had been illicitly taken. Perfitt gave his occupational title as ‘smith’.Footnote ⁵⁰ In May of 1686, Thomas Mercer, a Lancashire bricklayer, drove his sheep to and from the West Derby brecks.Footnote ⁵¹ Such examples of mismatches between titles and tasks are not difficult to find, but to what extent, if any, are they illustrative of wider patterns in the relationship between the two? What proportion of an individual’s work involved tasks unrelated to their principal occupational descriptor? One way to examine this is to construct the work repertoires for all men within specific occupational groupings, as shown in Table 2.15.Footnote ⁵²

Table 2.15Male repertoires by social/occupational status

Table 2.15Long description

The table depicts occupational task distributions across different social classes. The first column indicates the task category, followed by columns for average male repertoire in percentage, and repertoire percentages for gentry, professions, yeomen, artisans, husbandmen, labourers, and servants.

• For agriculture and land, the corresponding values are 31, 25, 14.5, 35.8, 13.1, 46.9, 36.5, 42.2.

• For carework, the corresponding values are 2.8, 9.7, 40.2, 5.1, 1.6, 2.1, 2.5, 2.2.

• For commerce, the corresponding values are 22.4, 18.1, 13.7, 29.8, 28.8, 18.9, 14.2, 7.2.

• For crafts and construction, the corresponding values are 10.6, 1.4, 0.9, 5.1, 27.4, 5.4, 9.6, 7.6.

• For food processing, the corresponding values are 6.8, 2.8, 0, 1.9, 8.3, 6.4, 9.3, 10.1.

• For housework, the corresponding values are 3.2, 5.6, 3.4, 1.4, 1.8, 1.7, 2, 5.2.

• For management, the corresponding values are 6.8, 22.2, 16.2, 10.2, 6.7, 4.7, 4.7, 5.2.

• For transport, the corresponding values are 14.8, 9.7, 5.1, 8.8, 11.3, 13.1, 20, 18.9.

• For other, the corresponding values are 1.6, 5.6, 6.0, 1.9, 1, 0.7, 1.0, 1.3.

• For total percentages, the corresponding values are 100, 100.1, 100, 100, 100, 99.9, 99.8, 99.9.

• For total tasks, the corresponding values are 6,964, 72, 117, 215, 1,267, 829, 591, 445.

The results for specific groups are discussed at various points throughout this section, but one clear overall pattern is the relative diversity of work repertoires for all types of occupation; for no occupational group is there one category that accounts for more than half of their work tasks. That said, there are clearly meaningful degrees of specialisation. The work of the professions, most of which are medical professionals such as physicians and apothecaries, has carework as by far the largest category at 40.2 per cent. The work of husbandmen is dominated, as would be expected, by agricultural tasks, which account for 46.9 per cent of their repertoire. Interestingly, the work of artisans is not as heavily concentrated in the category of crafts and construction, which accounts for only just over a quarter of their work tasks. Here the importance of tertiary tasks to those with secondary sector occupational titles is evident; commerce is in fact the largest category of activity for artisans, and transport also accounts for more than 10 per cent. Those working in the secondary sector spent a considerable proportion of their working lives buying and selling materials and goods, and fetching and carrying them between suppliers, employers, and customers. The tiler John Laurence of York, for instance, bought bricks, stone, and lime in 1594, while the plumber John Spurrel carried lead on horseback from Overstrand to Southrepps in Norfolk in September 1672.Footnote ⁵³

This is further illustrated by looking at the subcategories of work performed by some specific artisans. For butchers, 21 per cent of tasks recorded fall in the subcategory of butchery, but a further 28 per cent come from the buy subcategory, and 20 per cent from the sell subcategory, making commerce by far the largest of their work categories.Footnote ⁵⁴ This might be expected from more retail-oriented occupations such as butcher or baker, but commerce and transport are also well represented in the task profiles of manufacturing-oriented occupations. The proportion of commerce for shoemakers is 28 per cent, whereas the clothes and shoes subcategory, which includes shoemaking itself, accounts for 19 per cent.Footnote ⁵⁵ Interestingly, fieldwork represents a larger subcategory than shoemaking, at 21 per cent. Recently, Keibek and Shaw-Taylor have challenged the prevalence of by-employment in early modern England, particularly questioning the extent to which secondary sector workers were also engaged in agricultural work which would belong in the primary sector.Footnote ⁵⁶ For artisans as a whole, 13.1 per cent of all tasks recorded fall within the agriculture category, a not insignificant proportion of their work profile. This total conceals some telling differences between artisanal groups, however. Table 2.16 shows the overall work repertoires of four artisanal occupations frequently recorded as undertaking agricultural tasks. An additional row at the bottom of the table tallies all tasks possibly related to that occupation; so for weavers, this includes buying and transporting yarn, for instance, in addition to weaving.

Table 2.16Work repertoires of certain artisans

^a All ‘begging’

^b 4 of 5 ‘begging’

^c This includes tasks from all categories that might be considered related to that occupation: for instances, a weaver buying or selling yarn, or transporting cloth.

Table 2.16Long description

The table depicts task distributions across four craft professions like weavers, shoemakers, and carpenters. The first column indicates the task category, followed by columns of data for weaver tasks, tailor tasks, shoemaker tasks, and carpenter tasks.

• For agriculture and land, the corresponding values are 23, 23, 14, and 15.

• For carework, the corresponding values are 2, 6, 0, and 2.

• For commerce, the corresponding values are 29, 12, 16, and 14.

• For crafts and construction, the corresponding values are 29, 43, 11, and 57.

• For food processing, the corresponding values are 9, 1, 0, and 5.

• For housework, the corresponding values are 3, 2, 3, and 3.

• For management, the corresponding values are 10, 12, 7, and 7.

• For transport, the corresponding values are 23, 11, 7, and 5.

• For other, the corresponding values are 3, 5, 0, and 0.

• For total tasks, the corresponding values are 131, 115, 57, and 108.

• For tasks related to craft, the corresponding values are 49, 57, 22, and 66.

This analysis shows a spectrum of by-employment ordered along socio-economic lines. Traditionally low-income manufacturers, such as weavers, tailors, and shoemakers, were more likely to participate in agricultural work than the average artisan. Moreover, the poorer the primary occupation, the more likely individuals were to pursue work outside of and unrelated to it; weavers and shoemakers devoted less time to their primary occupations than did carpenters, who had access to better-paid construction work. Keibek and Shaw-Taylor have argued that by-employment was more common among relatively wealthy individuals, since the probate inventories of lower-income workers show little material sign of multiple occupations. From this they extrapolate that impoverished workers were more likely to rely on a single occupation, which mutes the overall effect of by-employment on measurements of sectoral distribution using occupational data, as discussed in Section 2.4. Yet our results indicate the opposite; poorer workers in the secondary sector, like weavers, shoemakers, and tailors, were more likely to seek work where they could find it. Much of this work might have been done ‘for another’ – 33 per cent of tasks performed by all artisans were recorded as such – and therefore often using others’ equipment or resources, materials which would not show up in the actor’s probate inventory.

If ‘out of sector’ work was only undertaken by artisans occasionally and seasonally, during the harvest, it may not have made a significant impact to the overall distribution of their work activity across different sectors of the economy.Footnote ⁵⁷ Certainly, our analysis of seasonality in Section 4.2 shows that craftsmen and tradesmen performed 50 per cent of their agricultural tasks during the harvest quarter. Yet, importantly, this distribution of labour was no more seasonal than that of husbandmen and yeomen, who did 47 per cent of their agricultural work in the same season. The share of artisans’ repertoires devoted to agriculture remained between 11 and 15 per cent during the other three quarters of the year, suggesting that such by-employment in agriculture was not merely a seasonal phenomenon. If husbandmen and yeomen were by-employed in craftwork to the same extent that artisans were in agriculture, then the overall effect on the economy and sectoral distribution would be limited.Footnote ⁵⁸ But as Table 2.15 shows, only 5.4 and 5.1 per cent of tasks undertaken by husbandmen and yeomen respectively fell within the crafts and construction category, as opposed to the 13.1 per cent of artisans’ tasks that were agricultural. This suggests that there was a significant degree of agricultural by-employment among artisans, and particularly among their poorest (and also most numerous) contingent. This helps to account for the sectoral distribution of tasks, with a much smaller secondary sector than one based on occupational descriptors, as discussed in Section 2.4. Those with secondary sector occupations undertook a significant number of tasks in both the primary and tertiary sectors.

Another way to think about sectoral distribution and the extent of specialisation is to investigate the proportion of work falling into certain categories that was performed by different occupational groups. In Table 2.17 the main occupational categories and work categories have both been grouped along broadly sectoral lines. It demonstrates that those with explicitly agricultural occupational descriptors performed 42.2 per cent of male agricultural work, with over half of agricultural tasks being undertaken by other occupational groups, including servants and labourers, some of whom would have specialised in agricultural work. Work tasks in secondary sector categories, those focused on crafts and food processing, were primarily performed by men with artisanal titles, at 54.4 per cent. The picture is very different for tertiary categories of work, where less than 10 per cent of male tasks were performed by individuals with a specific tertiary occupation. Instead, both agricultural workers and especially artisans undertook significant proportions of the work performed in the tertiary sector. If substantial proportions of primary and secondary sector men’s work were performed by specialists in those sectors, the same was not true of tertiary sector work. This offers further explanation as to why tertiary work is underrepresented by occupational data for this period, even when focusing on men’s work alone. Indeed, as the analysis here has shown, the relationship between occupational titles, work repertoires, and the sectoral distribution of work was far from straightforward.

Table 2.17Proportion of types of work performed by male occupational groups

Table 2.17Long description

The table depicts task distribution across social groups divided into four categories, agricultural tasks, crafts and food tasks, commerce and management tasks, and transport tasks. The first column has the professions, followed by values for agricultural tasks, crafts and food tasks, commerce and management tasks, and transport tasks. The four columns are further subdivided into number and percentage respectively.

1. 1. Agricultural Tasks:

   • For gentry and professions, the corresponding values are 35 and 2.8.

   • For artisans, the corresponding values are 166 and 13.4.

   • For yeomen, husbandmen, agricultural trades, the corresponding values are 523 and 42.2.

   • For labourers, the corresponding values are 216 and 17.4.

   • For servants and apprentices, the corresponding values are 203 and 16.4.

   • For tertiary commercial and transport trades, the corresponding values are 17 and 1.4.

   • For other miscellaneous and multiple, the corresponding values are 78 and 6.3.

   • For total, the corresponding values are 1238 and 99.9.

2. 2. Crafts and Food Tasks:

   • For gentry and professions, the corresponding values are 4 and 0.5.

   • For artisans, the corresponding values are 452 and 54.4.

   • For yeomen, husbandmen, agricultural trades, the corresponding values are 123 and 14.8.

   • For labourers, the corresponding values are 112 and 13.5.

   • For servants and apprentices, the corresponding values are 83 and 10.

   • For tertiary commercial and transport trades, the corresponding values are 21 and 2.5.

   • For other miscellaneous and multiple, the corresponding values are 36 and 4.3.

   • For total, the corresponding values are 831 and 100.

3. 3. Commerce and Management Tasks:

   • For gentry and professions, the corresponding values are 64 and 5.7.

   • For artisans, the corresponding values are 450 and 40.3.

   • For yeomen, husbandmen, agricultural trades, the corresponding values are 317 and 28.4.

   • For labourers, the corresponding values are 112 and 10.

   • For servants and apprentices, the corresponding values are 68 and 6.1.

   • For tertiary commercial and transport trades, the corresponding values are 81 and 7.2.

   • For other miscellaneous and multiple, the corresponding values are 26 and 2.3.

   • For total, the corresponding values are 1118 and 100.

4. 4. Transport Tasks:

   • For gentry and professions, the corresponding values are 13 and 2.3.

   • For artisans, the corresponding values are 143 and 25.4.

   • For yeomen, husbandmen, agricultural trades, the corresponding values are 150 and 26.6.

   • For labourers, the corresponding values are 118 and 21.

   • For servants and apprentices, the corresponding values are 88 and 15.6.

   • For tertiary commercial and transport trades, the corresponding values are 30 and 5.3.

   • For other miscellaneous and multiple, the corresponding values are 21 and 3.7.

   • For total, the corresponding values are 563 and 99.9.

If cross-referencing occupational titles and tasks performed can inform debates about the overall shape of the early modern English economy, it can also help historians to investigate the work undertaken by those with more opaque descriptors, which were particularly prevalent in the early modern period. From ‘gentleman’ and ‘yeoman’ to ‘servant’ and ‘labourer’, preambles often accorded people titles that said more about status or employment relations than the type of work an individual actually did. This is certainly true in the case of those women who were described as ‘wife’, a term that historians have argued might be seen as a form of occupational descriptor as well as a marital status, but which reveals little about the specifics of the work involved.Footnote ⁵⁹ Section 2.3 investigated the work repertoires of married women, but we can also use our data to examine the proportion of tasks undertaken by different groups of wives. In some instances, preambles recorded the occupational descriptor of a woman’s husband, and these can be used to reconstruct repertoires for the wives of husbandmen, artisans, and labourers.Footnote ⁶⁰ This is not to suggest that a husband’s occupational title can be readily extended to his wife; rather, it allows the extent of any correlation between the work of husbands and wives to be examined. This has been a subject of debate, especially in relation to the wives of artisans, where some historians have argued that the work done by artisans and their wives was very different, and others have suggested it was more complementary, with different but related tasks performed side by side.Footnote ⁶¹ It could of course vary from case to case. Amy Erickson suggested that in eighteenth-century London, it was more common for couples in the middling ranks to work in the same trade but unusual at lower socio-economic levels; even in the middling ranks, however, no more than half of such couples worked together or in related occupations.Footnote ⁶²

As Table 2.18 shows, wives of husbandmen did not consistently work side by side with their husbands in agriculture. Instead, they had a varied work repertoire, where commerce was the largest category, and much of the rest of their work was evenly distributed across the categories of agriculture, carework, housework, and transport. The only two subcategories for the wives of husbandmen that accounted for more than 10 per cent of their activities were buy, at 21 per cent, and carry goods, at 12 per cent, suggesting that rural wives took on considerable responsibility for the commercial activities of farming households, and that their work was particularly diverse. The work of wives of labourers appears to be similarly diverse, with commerce again dominating, but craft and construction taking a more prominent role. It is hard to say much about patterns at the subcategory level here, due to the small sample, but the differences between the wives of husbandmen and of labourers were limited.

Table 2.18Work repertoires of wives of certain occupations

Table 2.18Long description

The table depicts task distribution percentages for women across different occupational categories. The first column indicates the task type, followed by columns showing, average female repertoire, wife of husbandman repertoire, wife of artisan repertoire, and wife of labourer repertoire all in percentage.

• For Agriculture and land, the corresponding values are 17.6, 14.6, 9.5, 12.

• For Carework, the corresponding values are 13.7, 13.6, 6.3, 14.7.

• For Commerce, the corresponding values are 20.7, 29.1, 27.8, 33.3.

• For Crafts and construction, the corresponding values are 7.7, 9.7, 12., 12.

• For Food processing, the corresponding values are 4.9, 3.9, 5.7, 1.3.

• For Housework, the corresponding values are 19.5, 10.7, 18.4, 14.7.

• For Management, the corresponding values are 6.4, 3.9, 12, 4.

• For Transport, the corresponding values are 7.7, 14.6, 8.2, 8.

• For Other, the corresponding values are 1.8, 0, 0, 0.

• For Total percentages, the corresponding values are 100.0, 100.1, 99.9, 100.

• For Total tasks, the corresponding values are 2686, 103, 158, 75.

Wives of artisans performed less work in agriculture than wives of husbandmen, and more work in crafts and construction, as might be expected. They also played a key role in the commercial activities of their household, with both buying and selling appearing in their top three subcategories, alongside textile production. Management activities also formed a significant proportion of their work repertoires. This could imply that the wives of artisans often did have a close working relationship with their husbands, helping to manage a household business, selling goods it produced, and dealing with associated payments and finances, although it could also be evidence that they were running enterprises of their own that required these tasks. This can be examined further by analysing whether the tasks recorded for the wives of artisans in any way matched the occupation of their husband. Here this is defined as belonging in the same subcategory as their husband’s occupation, or fetching, selling, or buying goods related to their husband’s occupation. This shows a match in only 23 per cent of cases. Closer examination shows that these were mostly instances of the wives of textilemakers, and that they carried out the same textile and clothing work which women were associated with more broadly, such as spinning, and thus were not especially distinctive to wives whose husband worked in that trade. As such, we cannot assume they were directly collaborating with their husbands when doing such tasks. Other examples of ‘matches’ were the wives of butchers or bakers selling meat and bread, showing that there was potential for commercial partnership between wives and their artisan husbands. For instance, at a Monday market in July 1631, the blacksmith William Foster and his wife Christian together ‘were standing to sell their commodities’, including nails, at a stall in Spilsby, Lincolnshire.Footnote ⁶³ A close working relationship between married couples may have been more prominent in some trades than others, but the work-task data suggests it was not widespread. In the remaining 77 per cent of tasks undertaken by artisan’s wives, there was no clear correlation between the task undertaken and the husband’s occupation.

A man’s occupational title cannot be taken to show what work his wife undertook. In some cases, the descriptors accorded to men did not provide much indication of the tasks their own working life revolved around either. Gentleman and yeoman are both titles that say as much about social status as they do about work, but by reconstructing the work repertoires of each, it is possible to explore the types of tasks they were likely to perform. As Table 2.15 showed, the work of the gentry was more heavily concentrated in managerial work than it was for any other occupational group, and involvement in secondary sector work was extremely rare. Yeomen performed more agricultural tasks than their social superiors, but their working lives also involved high proportions of management and commercial activities, and fewer transport and food processing tasks than husbandmen and labourers: they were less likely to get their hands and boots dirty. Historians have found it particularly difficult to reconstruct the working lives of labourers, as wage accounts record only their paid work and probate inventories are biased towards wealthier labourers.Footnote ⁶⁴ As shown in Table 2.15, their repertoire suggests that the work of labourers was relatively diverse. Agriculture was the largest category, at 36.5 per cent, but this was not as high a proportion as it was for husbandmen. Transport was much higher for labourers, at 20.0 per cent, than for any other occupation, suggesting that the low-skill, time-consuming work of moving goods in this society often fell upon the shoulders of this group. This is confirmed by Table 2.19, which shows the top subcategories of the labourer’s repertoire. It also shows that labourers were heavily involved in buying goods, which may well reflect their dependence on the market because they lacked their own resources to grow or manufacture the products needed to run a household.

Table 2.19Largest subcategories (25+ tasks) for labourers, husbandmen, and male servants

Table 2.19Long description

The table depicts men as labourers, husbandmen, and servants. The first column is for labourers with the corresponding number of tasks, the second column represents husbandmen with the corresponding number of tasks, and the third column indicates men as servants with their corresponding number of tasks.

1. 1. Men as labourers and the number of tasks completed include:

   • At fieldwork, tasks completed are 57.

   • At animal husbandry, tasks completed are 49.

   • In buying, tasks completed are 47.

   • At carting, tasks completed are 36.

   • At farm transport, tasks completed are 34.

   • To carry goods, tasks completed are 33.

   • For selling, tasks completed are 29.

   • At threshing, tasks completed are 28.

2. 2. Men as husbandmen and the number of tasks completed include:

   • At fieldwork, tasks completed are 152.

   • At animal husbandry, tasks completed are 104.

   • In buying, tasks completed are 80.

   • At farm transport, tasks completed are 69.

   • To carry goods, tasks completed are 46.

   • For selling, tasks completed are 55.

   • At butchery, tasks completed are 26.

3. 3. Men as servants and the number of tasks completed include:

   • At animal husbandry, tasks completed are 86.

   • In fieldwork, tasks completed are 48.

   • At farm transport, tasks completed are 29.

   • At carting, tasks completed are 28.

   • To carry goods, tasks completed are 26.

The profile of labourers’ work can usefully be compared to that done by husbandmen, two groups that historians have argued were growing increasingly convergent in the seventeenth century, with many of the latter effectively living a similar existence to labourers despite claiming a higher status title.Footnote ⁶⁵ At the level of both categories and subcategories, labourers and husbandmen performed a similar set of tasks: agriculture, commerce, and transport were the top three categories for both. However, the work of husbandmen was much more heavily concentrated in the first of these categories. Indeed, as Table 2.19 shows, whilst both groups had the same top three subcategories, for husbandmen the gap between their top two categories and their other main subcategories was significantly wider than it was for labourers. Where the husbandman’s working life was dominated by agricultural activity, the labourer was required to be more of a generalist to make ends meet.Footnote ⁶⁶

Servants are another group for whom it is difficult to recover the precise nature of their work, and a repertoire analysis can be applied here too for both male and female servants, as shown in Table 2.20. For male servants, agriculture emerges as the largest category at 42.2 per cent. Crafts and construction account for only 7.6 per cent, and housework and carework combined just 7.2 per cent, which suggests that most male servants were servants in husbandry rather than domestic servants or assistants to craftsmen. Food processing accounts for 10.1 per cent of tasks, but a quarter of these were threshing, which for PST purposes would be seen as a primary sector activity. Like labourers, male servants engaged in a significant number of transport tasks, again suggesting that this type of work was the lot of the lowest status workers.Footnote ⁶⁷ Their subordinate position within households resulted in them engaging in far fewer commercial transactions than husbandmen or labourers. Within the category of agriculture male servants concentrated in particular on animal husbandry. Feeding and checking on livestock, and moving them around the farm, were routine tasks often delegated to male servants.

Table 2.20Work repertoires for servants

Table 2.20Long description

The table presents a comparative analysis of labor documentation across different historical record sources. The first column identifies temporal categories, followed by five data columns showing: the percentage of tasks recorded in quarter sessions, church courts, and coroners’ reports, total task counts for each category; and their percentage representation within the complete dataset. The data reveals how task documentation varies across different archival sources and temporal classifications.

• For agriculture and land, the corresponding values are 17.6, 21.6, and 42.2.

• For care work, the corresponding values are 13.7, 8.9, and 2.2.

• For commerce, the corresponding values are 20.7, 8.2, and 7.2.

• For crafts and construction, the corresponding values are 7.7, 6.9, and 7.6.

• For food processing, the corresponding values are 4.9, 5.5, and 10.1.

• For housework, the corresponding values are 19.5, 37.5, and 5.2.

• For management, the corresponding values are 6.4, 2.7, and 5.2.

• For transport, the corresponding values are 7.7, 7.6, and 18.9.

• For other, the corresponding values are 1.8, 1, and 1.3.

• The numerical values for total are 100, 99.9, and 99.9.

• The total tasks in all were 2686, 291, and 445.

Female servants are one of the few female occupational groups for which a repertoire can be reconstructed, as we have 291 tasks performed by women described as servants.Footnote ⁶⁸ By comparing Table 2.20 with Tables 2.4 and 2.8, it is evident that the work of female servants was similar to that of young and never-married women; this is expected, given the considerable overlap between these three characteristics. Female servants did a higher proportion of their tasks in agriculture compared to the average female repertoire, and also less commerce, carework, and management than other women. Their largest category of work by some distance was housework. Table 2.20 also shows that male and female servants had quite distinctive work profiles, in large part due to the considerable discrepancy in the amount of housework undertaken. And whilst agriculture was a top two category for both, the types of agriculture each engaged in differed. Male servants’ top two subcategories were animal husbandry and fieldwork; milking was the largest agricultural subcategory for female servants, whose top three subcategories overall were food and drink provision (35 tasks), milking (24), and laundry (21). Undoubtedly, the work of female and male servants overlapped, especially during the harvest, but overall they concentrated their time on quite different tasks. These findings for female servants also present an interesting contrast to some of the existing literature, which has used sources such as wage assessments where dairying is the most commonly mentioned form of work, and farm books like Robert Loder’s and Henry Best’s, which list brewing, baking, and malting alongside washing and milking as key tasks for female servants.Footnote ⁶⁹ Of these, only washing and milking appear in our top 10 subcategories for this occupational group. It may be the case that these more specialist work activities such as dairying, brewing, baking, and malting were undertaken by fewer households than more general tasks like milking, washing, and food preparation, and that our methodology is capturing the more common tasks, those that servants would have performed in almost every household in which they were employed.Footnote ⁷⁰

One of the things that marked servants out from other categories of worker was the extent to which they performed tasks ‘for another’. For male servants the proportion was 76.4 per cent of all their tasks, and for female servants 80 per cent. This indicates the extent to which service involved giving almost all your working time to your employer. The average for all male tasks in the dataset was 35 per cent and for all female tasks 36 per cent. As Table 2.21 shows, most male occupations fell within a range of roughly 10 per cent either side of the average, suggesting that the proportion of ‘for another’ work undertaken did vary by occupation in this period but not dramatically for those not in service. The other exception was the professions, at 55.6 per cent, where many of our examples relate to the provision of healthcare for people outside of the household, and in general professionals specialised in providing services to others. The surprise here is that the work undertaken by labourers was not more commonly done ‘for another’ outside of their own household than was the case for other men. Whilst our ‘for another’ category is only loosely correlated to all income-generating work, this does suggest that the labourers in our sample were not wholly absorbed in working for wages and still engaged in a considerable amount of subsistence-oriented work. In this respect, they again look to have been a poor relation of, rather than a different species to, the average husbandman.Footnote ⁷¹

2.6 Conclusion

This chapter has demonstrated the more detailed approach to work enabled by work-task data. Exploration of work by age group shows how women’s work varied significantly over the lifecycle in a way that men’s work did not. This finding is reinforced by variations in women’s work according to marital status. Most significantly, young, unmarried women did the highest proportion of housework, while responsibility for carework increased with age, and reached its highest proportion among widows. Married women’s work repertoires were characterised by high proportions of commerce and management, suggesting the responsibilities and opportunities that came with being joint head of household.

Evidence of men’s occupational descriptors shows that the distribution of occupations of men carrying out work in the work-task dataset differed little from that of occupations calculated by Keibek from probate documents. However, when work tasks are arranged to show sectoral distribution, the pattern is very different from that shown by male occupations. The tertiary sector was much larger, and the primary and secondary sectors smaller. This was not only because much of women’s work falls into the tertiary sector, but because occupations under-record men’s tertiary sector work. Both primary and secondary sector occupations required large quantities of work transporting and marketing goods.

Work tasks allow us to look inside occupations and see how they were constituted of different tasks. This shows the prevalence of by-employment, with men with all types of occupations engaging in agriculture. By-employment was particularly important for less wealthy artisans, who worked at a range of activities to make ends meet. It also allows the graded differences in agriculture-related work undertaken by men of different levels of wealth and status, from gentlemen and yeomen, through to husbandmen, labourers, and servants, to be viewed with clarity. Wealthier agriculturalists did more management and commerce; wage workers did more transport and threshing; those in the middle did more tasks closely related to agriculture. Differences in women’s agricultural work are also evident, with young women and female servants doing a higher proportion of agricultural work, particularly milking, while married women had a significant place in the harvest workforce.

Work ‘for another’ provides another important dimension with which to analyse early modern work tasks and reveals some surprises. Carework was the most likely category of work to be undertaken either for pay or for people outside the household. Housework was almost as often ‘for another’ as agricultural work. The lives of servants were dominated by work ‘for another’, but other occupations combined ‘for another’ work with varied forms of production and subsistence work. These themes form the foundation on which the analysis in the following chapters is based.