For society, people’s work is important because it contributes to macroeconomic development and growth. For individuals and households, work is important for subsistence and well-being, but the need to work also determines their everyday practices, routines, and time use, as well as their position, status, and identity. To truly understand and appreciate work in its broadest possible context, we should look at all forms of work, paid and unpaid, “free” and “unfree”, for the market, and for the household. We should consider how and where people are engaged in work activities, how work is organized, and the labour relations within which men, women, and children perform their work. However, as many scholars have remarked, conventional sources, such as occupational statistics, tend to give a skewed or incomplete picture of the actual work done by men, women, and children. In addition, the conventional definition of work as specialized, full-time occupations ignores the many combinations of work activities in which people are often engaged. A focus on occupations also overlooks much of the work done by women and children, in and outside the home, often unpaid. We need to find new sources, use well-known sources in new ways, think carefully about how to conceptualize the study object, and develop more sensitive methodologies to obtain a more comprehensive overview of what work entails.

As Jane Humphries and Carmen Sarasúa recently pointed out, several myths are associated with women’s participation in the labour force, two of which are of particular relevance to this Special Theme. The first is that a high female labour force participation rate is an entirely new phenomenon; the second is that women’s participation rates have always been considerably lower than men’s.Footnote ¹ To the extent that these myths are taken to be truths, they produce highly erroneous narratives about the trajectory of women’s work from early modern to modern society. In recent decades, however, much research has been done on women’s work, not only for the period of industrialization but also for the early modern period.Footnote ² Many of these studies show that women performed more work than often assumed, both paid and unpaid, that they carried out a broad spectrum of work activities, that they were not confined to indoor work, and that they were particularly active in textile production and trade.Footnote ³ While care for small children was usually the responsibility of women, it is nevertheless misleading to claim that married women’s work opportunities were significantly restricted by this responsibility. Instead, several studies suggest that the status of being married (or previously married) improved women’s opportunities to support themselves and contribute to the economy.Footnote ⁴ Other studies have noted that as “mistresses” (female heads of households), women could exercise authority over others – and that they were expected to do so.Footnote ⁵ By contrast, men were not always and not necessarily the main supporters of the household; they could be temporarily unemployed because of shifting seasonal demand in the labour market, they could be unable to head and support the household because of low public esteem, or they could be ill.Footnote ⁶ Not least, men could be absent, a common situation in the early modern period because of warfare, labour migration, and global trade. Consequently, far from all households were headed by a married couple.Footnote ⁷

Recent studies provide sufficient empirical evidence to demonstrate that some ideas about women’s and men’s work in the past are, as Humphries and Sarasúa put it, mere myths. It is incorrect to describe the high level of women’s involvement in paid work as a modern phenomenon, and a gross oversimplification to claim that men have always worked much more than women. These insights have consequences not only for how we think of the quantitative importance of women’s and men’s contributions to the economy, but also for how we think about power relations in households and in society more broadly. Even if religion and ideology prescribed that male heads of households were the ones who should wield power, this rule must have played out differently in practice, depending on how many men there were around. Therefore, these insights prompt historians to consider variation across time and space, and to develop even more fine-grained methods for analysing the sources. An early example of this effort was the Feminist Economics symposium “Off the Record” edited by Jane Humphries and Carmen Sarasúa.Footnote ⁸ One of the methods presented in this context is the triangulation of census data with other sources used by Cristina Borderías for the 1920 Labour survey and by Carmen Sarasúa and Ricardo Hernández for the Spanish Cadaster of Ensenada (1750s Property Survey). In both cases, the method leads to better data on women’s labour force participation.Footnote ⁹ In the same vein, Lotta Vikström has shown that triangulations between the census and local newspapers, trade directories, and business registers give a far better picture of the multi-occupational and part-time work of urban women in the late nineteenth-century Swedish town of Sundsvall.Footnote ¹⁰ In fact, triangulation of sources forms the basis of the productive and widely used micro-history approach that explicitly turns qualitative evidence into quantitative evidence by linking qualitative descriptions, for example work activities mentioned in court records, to parish registers, censuses, and tax registers, as Carus and Ogilvie have demonstrated.Footnote ¹¹

Historians have also developed other methods to capture the work of women and men. One approach has been to calculate women’s labour input based on what is known about demand. An early example is Kirsi Vainio-Korhonen’s study of how the city of Turku was supplied with clothing and footwear around 1800. Here, the size of the population is used to estimate the number of (anonymous) female producers who supplemented the officially registered production by men. Another example is the calculation by Ariadne Schmidt and Elise van Nederveen Meerkerk of women’s economic contributions to the Dutch textile industry based on the estimated need for yarn.Footnote ¹² Sara Horrell and Jane Humphries were among the first to analyse household budgets to calculate the economic contributions of women, men, and children in the first half of the nineteenth century, and Humphries has also used the potential of memoirs to delve deeper into children’s work.Footnote ¹³

This Special Theme continues this quest for better data on women’s and men’s work in the past. It originates in a workshop organized in 2023 by Karin Hofmeester at the International Institute of Social History in Amsterdam. The workshop brought together historians studying various aspects of work in the past, and made clear that new methodologies, sources, and digital tools are changing the scene for labour history. As a follow-up to the workshop, two sessions were organized at the European Social Science History Conference in Leiden in March 2025. The Special Theme showcases some of the results presented at these meetings, and focuses on experiments where new combinations of sources and methods are used to provide more complete and, when possible, quantifiable data on work done by women, men, and children in Europe and Asia, from the seventeenth to the late nineteenth century. All three articles apply a broad and encompassing definition of “work” that includes production as well as services, paid as well as unpaid work, work for the market as well as for the household, and free and unfree work. Arguably, this broad definition of “work” is close to the way people understood work in the early modern period – or today, for that matter – that is, as “worthy efforts” with the purpose of securing well-being. This definition differs from the way in which “work” is often defined by economists.Footnote ¹⁴

Court cases are at the centre of the triangulations described in the three articles that make up this Special Theme. These cases are analysed with the help of a new tool in the historians’ toolbox: the verb-oriented method. This method requires the scholar to identify and extract information about work activities that were, according to the source, carried out by specific individuals. Activities that cannot be linked to specific individuals, or that are presented as ideals (for instance, “embroidery is a fitting task for a young woman”) are not included in the data selection process. This is because the rationale behind the method is to come as close as possible to people’s actual everyday work practices.Footnote ¹⁵ The method was inspired by Sheilagh Ogilvie’s pioneering study of a community in early modern Württemberg and further developed in the Swedish Gender and Work (GaW) project, which coined the name of the method.Footnote ¹⁶ Inspired by Ogilvie and the GaW project, Jane Whittle and Mark Hailwood developed the method further for English sources using what they term the “task-based method”.Footnote ¹⁷ Danielle van den Heuvel and her research team have applied the verb-oriented method to Dutch witness depositions, extending its usefulness by observing not only activities but also their spatial characteristics. More recently, a large project that uses the verb-oriented method to study work in early modern Italy has started under the leadership of Andrea Caracausi.Footnote ¹⁸ The GaW group has also recruited two researchers who will apply the verb-oriented method to Portuguese and Polish sources.Footnote ¹⁹

The articles that follow focus on the early modern period and the nineteenth century. This was a period when many parts of Europe experienced great or even spectacular economic growth, while many other parts of the world were drawn into relations of dependence. At the same time, many people lost their connection, or right, to land and had to enter into new types of labour relations, not least in the form of commodified labour, to make a living. To explain these processes, it is vital not to repeat the two myths Humphries and Sarasúa identified. It is not easy, however, to provide better estimates of women’s work, which is why the following articles are presented as experimental snapshots of work in the Swedish region of Västmanland (1880), its capital Västerås (1820), colonial Batavia (c.1635), and Japanese Edo (early nineteenth century). Work activities performed by individuals as captured in court cases were triangulated with an early census (Västmanland), early population statistics (Västerås), and with location data (Batavia), after which a second layer of analysis was applied.

The first article deals with one of the myths mentioned above. Did men work considerably more than women, as nineteenth-century censuses would have us believe, and if not, what work did women do? In their article, Jonas Lindström and Maria Ågren use for the first time the verb-oriented method for the post-1800 period to triangulate occupational descriptors from the 1880 census of the mid-Swedish region of Västmanland with the qualitative information on work activities found in court records from that same region. The main goal of this methodological experiment is to obtain better data on the occupational structure of the whole population, including women, as an indicator of economic development just before the onset of industrialization. As both sources mention names, they first used the verb phrases from the court records to check the reliability of the occupational descriptors in the census. The census provided occupational descriptors for only twenty per cent of the adult women, and mostly in vague and perfunctory terms, so the next step was to establish women’s work activity patterns and occupational structure and compare this with the occupational structure of men. Lindström and Ågren did this based on the proportion of work activities found in the court records and on the assumption that women worked as much as men. The triangulation and the methodologies for adjusting data in this particular “snapshot” result in new estimates for the proportion of women and men in the primary, secondary, and tertiary sectors, with a significantly higher proportion of women than men in the primary sector (which includes agriculture), and a larger tertiary sector than generally assumed. Within that sector, they find a larger-than-expected proportion of women, not just in care, food, and accommodation, but also in trade and credit. Lindström’s and Ågren’s methodological contribution to this Special Theme shows, first, that verb-phrase data can be used for estimating work-activity patterns of women and men in the late nineteenth century, and, second, that these data can be combined with data from a radically different source, such as the census, and in fact also be used to evaluate the latter.

The second article is related to the problem of understanding what is behind occupational descriptors. Did people have one full-time occupation only, or did everyone in fact engage in multiple employments? This is the topic of the second article, in which Karin Hofmeester, together with Maria Ågren and Jonas Lindström, evaluate another source with the help of verb-phrase data. Here, the town of Västerås in 1820 is the scene of the snapshot in which data from the population statistics in the Tabellverket is used to apply labour relations to the total population of the capital of Västmanland. These labour relations are defined theoretically as “for or with whom one works and under what rules”, and are indicative of the social and power relations determined by work. With the help of the verb-phrase data the occupational and status categories of the Tabellverket are “translated” into labour relations and these categories are then complemented with information on work activities not mentioned in the Tabellverket, namely, men’s other work activities – their by-employments – and women’s work activities, as found in the local court records. A key idea behind both the labour relations approach and the verb-oriented method is that people usually combine work activities rather than just having one, so they can have more than one labour relation. Unlike the 1880 census, the Tabellverket does not contain names, so no systematic and all-encompassing reliability check could be performed. However, as some individuals in the verb-phrase data belonged to a small occupational or social position category of the Tabellverket, they could be identified and their work activities were considered to be representative of the whole category. For most adult men, the (assumedly) main occupation was captured in the Tabellverket and so attributing a labour relation to it was relatively easy. For women’s work, especially married women’s work and for men’s secondary work activities, and thus secondary labour relations, the Tabellverket lacks information. Here, labour relations had to be reconstructed with the help of verb-phrase data. The result of this triangulation is a more plausible estimate of women’s work and labour relations in general, and of married women’s work in particular. Many “wives of” worked in trade (a sector underrepresented in the Tabellverket), or in food and accommodation services, often as self-employed. Many single women worked as wage earners doing care work on a commodified basis for working mistresses. Since the combination of both methods does not answer the question of the quantitative importance of by-employment and care and domestic work, the assessment of women’s labour relations is presented as a possible range, with an upper limit of eighty-seven per cent adult women whose first labour relation was commodified and a lower limit of seventy-two per cent.

The third article, by Danielle van den Heuvel, addresses the shortcomings of current forms of big data to describe work practices. It advocates a more variegated approach to the study of work by combining visual and textual sources, and by applying a space-time framework. It builds upon the systematic analysis of witness depositions as employed by the Freedom of the Streets project that is not restricted to work activities but extends to all types of activities, triangulating activities of individuals or groups of individuals with location and time. This means that different types of work and different groups of workers can be presented in a single comprehensive analysis, tied to a particular location (pinpointed on a georeferenced map) and time of the day. In her snapshot of colonial Batavia in 1653 describing a pre-crime scene, we meet adults, children, men and women, enslaved, wage-earning and self-employed, and we learn about their work practices, routines, time schedules, but also about the functioning of the city and its economy. This analysis is enhanced by applying the same snapshot method to pictorial evidence, as developed by the interdisciplinary team in the context of the cross-cultural comparison of cities with differing historic registration practices.Footnote ²⁰ As Van den Heuvel demonstrates by using contemporary sketches and paintings, analysing textual evidence and visual sources in tandem is essential to come to a fuller understanding of the wide range of work undertaken in the urban economies of the premodern period. Her article offers insights into the occurrence of combined work tasks, into temporal and spatial work routines, as well as into tasks such as care work that are commonly underreported in many textual sources. It concludes that the same sensitivity to fabrication and storytelling needs to be applied to the analysis of visual and textual sources, and encourages scholars to embrace the creativity and uncertainty inherent in anecdotal evidence to create better data for writing new histories of work.

To get a better picture of the work of women (and men and children), the research procedure behind the snapshots presented in the three articles in this Special Theme must be repeated and the results compared across time and place. Comparisons can then be made between regions with different economic contexts, for example the presence of a textile industry, the degree to which a region is cash-poor, and urban and rural cases also can be compared. These regions can be monitored over time, for example to see the effect of industrialization on female labour. The inclusion of colonial cities opens up the possibility to compare within and across colonial empires and to compare colonial and post-colonial work activities, labour relations, and the long-term effects of colonialism. Comparing the snapshots requires serial sources. Court cases are available in many regions and cover long swathes of time – which is precisely why they are in frequent use. The possibility of using serial pictural sources opens new exciting research vistas for areas where other sources might be less easily available. This makes these snapshots reproducible and the results comparable over time and space, with the help of the methods described in this Special Theme.

The history of work is marred by the fact that the meaning of “labour” or “work” changed with the arrival of modern society in a way that makes it difficult to make comparisons across time. There has been a shift from understanding work as any efforts that may secure continued living and well-being to seeing it as paid, full-time, specialized activities only. In this process, the work of some groups in society (notably women, but also others) and work in the form of multiple employments (which often means multiple labour relations) have become more difficult to grasp. The methods and sources presented in this Special Theme promise to help historians overcome these difficulties.