Reassessing Registers and Registration Practices

In 1742, a new tax was issued across the Northern Netherlands to raise funds for the War of the Austrian Succession. One result of this Personele Quotisatie tax was one of the most comprehensive lists of households across a wide range of communities throughout the provinces, including for major cities such as Amsterdam, Rotterdam, and Leiden. This was exceptional for its time and in Europe: in many areas, systematic overviews of the working populations came only in the nineteenth century, when censuses and population statistics became more common. The Amsterdam register, like its counterparts for other settlements throughout the country large and small, is one of the registrations most widely used by historians to reconstruct and analyse the city’s premodern economy. As it lists householders with their occupations, residential addresses, and indicators of wealth common at the time (such as annual income and numbers of servants and of horses and carriages), it provides unique insight into two important markers of the state of the economy: wealth distribution and occupational structure. Despite its well-known omissions, most notably the absence of the wealthiest householders and those of lower social means, its value for pre-statistical economic history is immense. The systematic overview of taxable wealth linked to occupations and residences offers insights into a variety of aspects of the preindustrial economy, and historians up until the present have used this source to uncover details about the real estate market, retail development, and, through registrations of domestics and coaches and horses per household, the size of sectors such as domestic service and transport.Footnote ¹²

The 1742 tax register, and comparable registers of similar status, have also been instrumental for the in-depth analyses of women’s work in the premodern Netherlands. In that context, I studied the Personele Quotisatie register to assess the role of women in the Amsterdam economy in the century after the so-called Dutch Golden Age. While providing the opportunity to analyse the occupational titles of male and female householders in the eighteenth-century city, the register simultaneously yielded intriguing questions on the work of household members who were not heads of household, as well as on gender differences in occupational titles and on the relationship between occupational titles and actual occupational activities. Let me highlight one example. Of 1,665 women in total, the register included no fewer than 769 women listed as rentiers.Footnote ¹³ This listing was typical for women, outnumbering all other types of occupations for women in this register, and, crucially, it was much less common for men, who, if they came from similar backgrounds and possessed comparable worth, were generally listed as merchants.

The historiography tends to directly oppose these male merchants to the female rentiers: active, entrepreneurial tradespeople are set against passive investors who had withdrawn from the world of commerce to draw an income from their wealth. Jan de Vries and Ad van der Woude’s classic and still widely used book on the preindustrial Dutch economy is typical in its characterization of these rentiers as coupon clippers.Footnote ¹⁴ However, a closer look at the actual activities undertaken by the rentiers and merchants included in the 1742 register reveals that such long-standing assumptions are largely based on semantics, bolstered by gendered notions of what work is or should be, rather than actual activities. The Amsterdam Bank of Exchange ledgers reveal (at least part of) the financial transactions of the merchants and rentiers included in the 1742 tax register and show considerable overlap in the size, type, and occurrences of their transactions between men and women, merchants and rentiers. Strikingly, the difference in both quantity and quality of the types of activities is not as large as one would assume based on the occupational titles, and regardless of whether they were listed as merchant or rentier both women and men operated in commodities as well as in financial trade and investments.Footnote ¹⁵

More recently, the observation that occupational titles often do not match the work activities individuals performed has been established for other areas of preindustrial Europe. Among the most prominent scholars investigating this discrepancy are those involved in the Uppsala-based Gender and Work project led by Maria Ågren. This group has been setting a new standard for the history of work through their development of a new methodology to uncover and analyse work practices (see their articles in this Special Theme). Their results have shown that, in early modern Sweden, too, occupational titles do not necessarily reflect work activities. In fact, their results show that, in rural Sweden, people not only combined various ways of making a living at the same time – at times, occupational titles and work activities did not even overlap. This discovery made it even clearer that an occupational title is a social marker rather than an accurate characterization of the work someone undertook. Occupational titles are constructions that should be carefully unpacked before their full meaning and validity are assessed. This is especially the case for women’s work, as the categories in the registers are found to be gendered and therefore essentially meaningless if one adopts such categories without interrogation. This insight resonates with observations made in the field of colonial history and with regard to the history of the subaltern. Here, historians have also encountered how labels applied by administrators from the past are often at odds with the complex historical realities underlying such terms.Footnote ¹⁶

What does the application of such methods yield? In the case of the Gender and Work project, profound new insights have emerged on the workings of the household economy through the identification of the ideal and the reality of the two-supporter model, thereby raising serious questions concerning the relevance of the rise of the male breadwinner as a historical reality. However, I want to turn to a related project to illustrate what I would call the power of the anecdotal.Footnote ¹⁷ In 2016, The Freedom of the Streets. Gender and Urban Space in Eurasia (FOSGUS) research project was established at the University of Amsterdam to investigate the long-standing issue of the gendering of urban space and the supposed withdrawal of women from city streets as a result of modernization. In this project, we asked simple but important questions that were largely unanswered at the project's outset: How did women move through cities? Where did they go, and when and with whom? How do patterns of mobility differ according to social status, wealth, and age? How does women’s movement compare to that of men? And finally, does female mobility change as cities enter the so-called modern era? We studied four cities in different parts of the world – Amsterdam, Edo, Batavia, and Berlin – to assess the impact of architecture and urban planning, economic structure, governance, and cultural constraints.Footnote ¹⁸

One of the biggest challenges faced by the Freedom of the Streets project was the charting of everyday movement and street use, such as when individuals moved between home and workplace, bought provisions, or engaged in activities such as sweeping the stoop in front of their house or approaching a neighbour for a casual chat. As Colin Pooley states, such mundane and ubiquitous practices are generally perceived as unremarkable, and thus are seldom recorded and leave little if any traces in the archives.Footnote ¹⁹ While I do not fundamentally disagree with Pooley, I would like to assert that we often assume that such practices cannot be found in the archives even though the evidence is there, at times even abundantly. Indeed, by applying a practice-oriented approach and by looking to sources previously (largely) neglected, suddenly, invisible phenomena of this sort become highly visible.

Inspired in its aims and methodology by the Gender and Work (GAW) project, our task started with meticulously reconstructing the street use by women and men, collecting observations on activities that took place in the street.Footnote ²⁰ As our remit was wider than the GAW project, we collected not just work and work-related practices but all types of activities. Moreover, as we wanted to pinpoint where certain activities took place and in what kind of spatial conditions, we developed a new methodology tailored to our specific project goals. Taking the verb-oriented method as a starting point, the new Snapshot Method, developed as part of the FOSGUS project, allows for the capturing of snapshots of everyday life, in a triangular constellation that connects activity – location – individual. These snapshots can be pinpointed on a georeferenced map, which not only allows the spatial analysis specifically required for the FOSGUS project, but also offers opportunities to analyse multiple activities that happen in the same location, either simultaneously or consecutively. As such, we can connect activities extracted from different sources, activities that would not have been so easily related otherwise. We can also study the relation between activity, time, and space and investigate the character of outside spaces throughout a day, a season, or a year, following Torsten Hagerstrand’s influential Time-Space Geography.Footnote ²¹

Our approach, similar to that of the GAW project, had us begin with the most commonly used materials for gathering such snapshots: witness depositions. We prioritized the incidental evidence concerning the circumstances that led to the incident for which the record was created. Our eyes were not on the crime or the confrontation between people but on what culprits, victims, and eyewitnesses were doing at the moment right before the incident took place: the “pre-crime scene”.Footnote ²² For our case studies, we expanded the source base to include a variety of materials that reveal, akin to what we find with court records, snapshots of everyday experience in the distant past, including diaries, travellers’ accounts, and crime reports. We also included visual materials that depict street scenes, such as cityscapes and illustrated city guides. The implications of this expansion (both positive and negative) are discussed below. However, our cross-cultural study of cities with different historic registration practices, whose archival collections vary in character and scope, meant that it was essential to widen the evidence base if we were to extract everyday practices on a large scale. More importantly, perhaps, our desire to capture as best as we could the snapshots of everyday ephemeral activities convinced us that we had to cast our nets as widely as possible. As we will see, our results justified this assumption, and they also yielded some important challenges for future research.

Turning Anecdotes into Snapshots: A Batavia Murder Case

To illustrate the FOSGUS approach and its insights, I will zoom in on one particular case: an enslaved man who ran amok in the city of Batavia in the mid-seventeenth century, as reported in the Dagregisters (Daily Registers). In these registers, Batavia’s colonial government reported all events they deemed significant to the Dutch East India Company’s (VOC) directors in the Dutch Republic. These registers, shipped regularly to Amsterdam from Batavia and other VOC trading posts, have survived for almost the entire VOC period. They contain reports on ship arrivals and on the appointments of officials, as well as on matters relating to the city’s population size. Cases such as the episode reported for 14 March 1653 are also documented.Footnote ²³ On that day, according to the report, an anonymous enslaved man went on a rampage through the city. A few days before, he had escaped his enslaver, a Batavia resident of Asian-Christian descent (customarily labelled a Mardijker) named Manuel Ferere. The enslaved man sought refuge outside the city walls, but on the morning of 14 March he re-entered the city and went on a killing spree that left five people (including himself) dead and six wounded.Footnote ²⁴ The route he took, the people he encountered, and those he hurt are all meticulously registered in the report about the incident, allowing us to precisely map his movements through the city as well as what took place on that early Friday morning in March.

Let us start with the route, shown in Figure 1.Footnote ²⁵ The runaway entered the city from the south at the Nieuwpoort, one of the city gates. He proceeded along the prestigious Tigers canal (Tygersgracht), which housed the city’s wealthiest residents, all the way to the Castle of Batavia, the seat of the VOC government. Here, he turned left, passing the Steenhouwersbrug, a bridge connecting the inner city to the land where the castle stood, then crossed the Herenstraat junction and found himself at the river bank. The river separated the initial walled Dutch settlement from the inner city’s newest expansion west of the river. Here, he wounded his first victim: the bargeman. There was an altercation, presumably over the fugitive’s desire to cross the river using the barge, and the bargeman suffered a knife wound under his armpit before pushing his attacker into the river and seizing the fugitive’s knife. After swimming across the river to the other bank, the runaway stole a meat cleaver from a Chinese meat vendor, the weapon he would soon use on his other victims. He came upon a Malay woman selling crackers near the Generaalstuin (General’s garden) on the corner of Utrechtsestraat and stabbed her under her right arm, but the intervention by a passerby, identified as master bricklayer Cornelis Bruyn, saved her life, forcing the runaway to quickly leave the scene. Continuing along the river, he turned a corner and passed a bridge near the bulwark Punt Grimbergen. Just beyond the Hof van Batavia inn, he fatally stabbed, first, a small Chinese child and then, two doors down, a Chinese man on the doorstep of the man’s house. A little further down the same block he killed another Chinese man. Turning a corner, he spotted a seated woman, a Chinese shopkeeper’s wife with a child in her lap. Violently pulling the child from her, he stabbed the woman in her legs and hips. His next victims were two enslaved women; first he attacked a woman owned by the Mardijker Thomé Svegedo, then he struck another enslaved woman with the meat cleaver on the corner of the Stadsbinnengracht and Spinhuisgracht canals. He encountered his final victim, the Mardijker Paulo Suaris, on the bridge crossing the Spinhuisgracht canal, and fatally stabbed Suaris in the back. The runaway continued moving until he arrived at the home of the citizen Nicolaes Snyder on the same canal, where several men were approaching in hopes of stopping him. On one side, armed with a pike, there was the army corporal Jan Jeronimusz Mentser, summoned by his sergeant from the bulwark of Punt Nassau; on the other side was a group of “Chinese and Blacks with sticks and bamboos” supported by the aforementioned Cornelis Bruyn, brandishing a knife. Not long afterwards, the escapee met his end: he died of the wounds inflicted upon him during a final fight in the marshy canal he had fled to.

Source of the underlying map: The British Library, London. Public domain.

Figure 1. Route of the perpetrator in the 1653 Batavia murder case, as projected on a plan of the city of Batavia, 1650.

When we turn to what we call the “pre-crime scene” and imagine walking the route taken by the runaway just minutes before he did so, a totally different scene reveals itself (Figure 2). At around 8 a.m. on that Friday in March 1653, the city of Batavia was bustling with activity. Near the Castle, and opposite the market, a ferryman (a) stood ready to shuttle passengers across the river. On its banks, a Chinese street vendor (b) just back from his door-to-door rounds selling his meats had now put down his pork-filled baskets and was cutting chops for his stall’s next customers with a parang, a long cutting knife with sharp edges on both sides commonly used by the Chinese. A little further south, a Malay woman named Djago (c) was also preparing food: she was busy baking snacks (“koeckjens te backen na deses landts wyse”) for sale on the go to passersby.Footnote ²⁶ A master bricklayer, Cornelis Bruyn, also known as Meester (Master) Kees (d), walked along the street. Somewhat further to the south, near the Hof van Batavia inn, a small child of Chinese descent, not yet five (e), was playing in the street. A couple of neighbours, all Chinese (f, g), passed the time on their doorsteps; one of them, the wife of the shopkeeper Banco (h), had a toddler on her lap. A little further on, a few enslaved women (i, j) were out walking, as was Paulo Suaris (k), a Mardijker man who walked along the inner canal (the Stadsbinnengracht). Across the canal, on the Nassau bulwark, two men stood guard (l, m).

Source of the underlying map: The British Library, London. Public domain.

Figure 2. The west side of the walled city of Batavia, showing the city’s early morning street economy, c.1653, as projected on a plan of Batavia, 1650.

The above account of an enslaved man who escaped his master and fled the city is one of many such episodes. Runaways fleeing slavery, both male and female, were a common feature in Batavia, as Eric Jones has pointed out, and inextricably linked to the fact that it was a slaveholding society. Accounts of their escapes, recaptures, and subsequent punishment are numerous in the city’s criminal records.Footnote ²⁷ Yet, the early date and the level of detail this case provides, on both the crime itself and the people affected by it, are extraordinary. The vivid record of the episode is probably due to the crime’s brutality, which was exceptional enough to have made local officers keen to carefully report the smallest details of the perpetrator, his victims, the bystanders, and all their activities. As such, it gives us unique insight into the everyday goings-on in Batavia during the early 1650s, a decade relatively underexplored in the historical literature on the city, which tends to focus either on its early days (1610s–1630s) or on the long eighteenth century.

In the mid-seventeenth century, the city was experiencing rare stability. Its population’s size and composition seem to have been fairly constant. We lack exact population numbers for the 1650s: probably, the city’s relative stability meant that strict population control was deemed unnecessary, hence there is no paper trail of population statistics. We may assume that population numbers were somewhere between the 8,000 recorded for 1630 and the 17,000 for 1679, but the number is likely to have been closer to the lower 1630 estimate. We know that only from the late 1650s onwards was there growing concern about new groups of immigrants from various parts in the archipelago seeking to settle in the city, as well as a growth in the Mardijker and Chinese populations. In the early 1650s, when we are offered a glimpse of Batavia’s street life through our murder case, the city’s population size and composition were relatively stable, made up of large proportions of Chinese, as well as enslaved people and Mardijkers; Europeans formed only a small, albeit privileged proportion of the population.Footnote ²⁸ The people encountered above therefore fit neatly into this picture: they form a mixture of people of European descent with VOC connections and associated privileges (the citizen Snyder, the guild master Bruyn, and the corporal and his sergeant), Mardijkers, who were official residents of the city as well as slaveholders (Ferere, Svegedo, and Suaris), male and female enslaved servants (the kaffer operating the ferry; the women who were killed), Chinese salespeople (Banco’s wife; the anonymous pork vendor), and some Chinese residents of the newly laid-out western part of the city.

Mapping the presence and activities of these individuals, as snapshots, onto a contemporary map of the city yields important insights into everyday working lives in the mid-seventeenth-century colonial city. Our knowledge of work practices in Batavia is still fairly limited and, due to the divided attention the topic has received, highly compartmentalized. We possess separate studies on Moor sailors and dockworkers, on Chinese merchants and peddlers, on enslaved workers in the household and in the barracks of the Ambachtskwartier, and on agricultural work, especially the sugar industry, in the Ommelanden districts outside the city.Footnote ²⁹ We have some knowledge, too, on how VOC officers spent their working days, as well as on officers of the church and charitable institutions.Footnote ³⁰ Apart from this fragmented knowledge of male labour practices, knowledge of women and children’s work is lacking. Admittedly, there is a well-established tradition of scholarship on women in Batavia, but for a long time its focus was largely on elite women who, in accounts of their lives and their roles in Batavian society, were depicted merely as pawns in marriage politics and, parroting the many contemporary descriptions of the city and its inhabitants, as idle women who did not lift a finger in the running of their households or the raising of their children.Footnote ³¹ Recently, the topic of women has received renewed attention, yielding more nuanced interpretations of the role of women in this colonial society. This wave of scholarship has also shown increased interest in women who were enslaved and worked in the city’s households.Footnote ³² The role of child labour remains largely ignored, as does the work of women who were not part of elite households either as mistresses or as enslaved servants. Strikingly, in our collection of snapshots extracted from the murder case, representatives of almost all these types of workers appear in one (extended) scene. Analysing this case using the FOSGUS approach offers us the chance to bring together different types of work and different groups of workers in a single comprehensive analysis, tied to a particular location and time of day. Let us turn to the insights the analysis yields and see how they may help enhance our understanding of working lives in a seventeenth-century colonial town.

Routes and Routines: Spatio-Temporal Aspects of Work

The first conspicuous aspect of our 1653 case is its revelation regarding how people of various status, gender, and age were to be found in the streets on an ordinary Friday morning. Such diversity may not seem especially remarkable, but given the persistence of the debate on the nature of segregation in the city of Batavia it is an important fact to observe.Footnote ³³ Going about their everyday activities, urban inhabitants of various types crossed paths and shared spaces with one another. Even if we ignore their ethnic designation, which surely captures only partly the identities of the people involved, the other information provided in this source allows us to distinguish adults and children, men and women, as well as enslaved, wage-earning, and self-employed workers. Another important aspect of this case and of what it reveals through a dissection of the scene into snapshots of working practices is its illumination of the effect of spatial features on work routines. The impact of urban form on the city economy is widely acknowledged in historical scholarship, and scholars have used street patterns derived from contemporary maps to digitally model historical traffic flows to explain, for example, spatial patterns of commercial development. Such analyses reveal how the urban form both facilitated and obstructed economic activities and how the spatial layout of a city was a major factor in the blossoming of particular areas that were well connected.Footnote ³⁴ Batavia’s rigid grid layout, it is argued, and its lack of bridges connecting its different parts obstructed movement between various areas of the city.Footnote ³⁵ As we will see below, while the city’s form dictated movement via particular routes, the city’s sections were not disconnected from one another, and its layout not only exerted negative effects but also created opportunities for small-scale enterprises. Finally, the case shows the impact on work of routines and time schedules enforced by a mixture of climatic conditions and institutional regulations. Given what we currently know about the nature of work in the premodern era, and the fact that, for many people, the working day consisted of many different, sometimes wide-ranging tasks, we can glean important insights into the functioning of premodern economies if we grasp what was done when.Footnote ³⁶

Mornings in Batavia started early, around 5 a.m. to 6 a.m., as in other areas near the equator where the temperatures from late morning through late afternoon rose to levels that many people found unbearable. The physician of Jan Pieterszoon Coen, Batavia’s first Governor-General, advised his patient not to work between the hours of noon and two. It is also reported that, in light of the high temperatures, the company’s workers were instructed to work only from 6 a.m. to 10 a.m. and again from 2 p.m. to 6 p.m. François Valentyn, who reported extensively on life in the East Indies, writes that a large part of the day, roughly between 9 a.m. and 5 p.m., was too hot to undertake any activities.Footnote ³⁷ The start of the day was announced by the ringing of the bell (three times) at City Hall at 5 a.m. or 5.30 a.m.Footnote ³⁸ The city gates were then opened by an officer called a wachtmeester, who, accompanied by six guards and a drummer, would first fetch the keys to the city gates at the bailiff’s residence, then proceed to the gates to open them; he and his company of men would then return the keys to the bailiff.Footnote ³⁹ After the gates had been opened, the inner city would be accessible to people from the suburbs and surrounding countryside. Our victims were targeted between 8.00 a.m. and 9.00 a.m., just before the temperature began rising. The Nieuwpoort gate that the killer first entered had been open for at least an hour when he came through. Our culprit did not attack any of his victims on the east side of the inner city, so we have no details on how many people he encountered and who they might have been. The city gates, it has been suggested, were bottlenecks in the flow of traffic, especially from the 1670s onwards, given the growth in the population of the suburbs and Ommelanden. The Nieuwpoort was the southern suburb’s central point of entry into the city, so it is unlikely that this man came through the gate alone, and almost surely there would have been people on the streets on the city’s eastern side.

As for the city’s western side, our case provides much more detailed information. Most of the enslaved man’s victims would have been working for some time when they were attacked; the crime report reveals that the pork vendor had already made his door-to-door rounds selling his cuts of meat before setting up his stall at the water’s edge. The soldiers at their watch posts on the bulwarks would have been at their stations since early morning.Footnote ⁴⁰ Djago, the Malay woman selling ready-made crackers for consumption on the go (for the “komende en gaande man”), had probably been at her spot for a little while, too. We learn from mid-nineteenth-century descriptions of city life in Batavia that, around seven in the morning, when the streets would get busier, gaarkoks (cooks) would set up stalls on street corners to sell rice served in pisang leaves to workers on the move.Footnote ⁴¹ It is likely that this hour marked a common start to the working day in the seventeenth century as well, as in 1620 it was remarked that most workers would not arrive at work before seven, even though the company preferred them to start an hour earlier.Footnote ⁴² By the time she had been stabbed, sometime after 8.00 a.m., Djago had probably already served plenty of customers. Her continued presence after what we may regard as Batavia’s rush hour, however, suggests that enough people were still passing her stand to make it worth her while to keep selling. Our case, in fact, gives several hints of such movements, illuminating the flow of morning traffic in the city.

The first indication of traffic flows is provided by the man who comes to her rescue and later assists in the arrest: Cornelis Bruyn. Reportedly, he was walking on Utrechtsestraat when Djago was attacked. From the 1650 land register, we know that Cornelis Bruyn owned a plot of land in this part of the city – the fifth house from the northwest corner, which, according to a 1652 map, was the site of small row houses.Footnote ⁴³ While ownership of land does not necessarily mean that one lived on a property on the plot, we may assume that this is where Cornelis lived, surrounded by neighbours of similar status who occupied dwellings of a similar size. To his left was Jan Springer, a skipper, who owned the plot on the corner, and to his right sequentially Jan de Moff, Jan Tamboer (a sergeant), Jan van Gorcom, and another skipper named Aert Plumen Moot. Most others occupying land in this block also bore European names and worked in shipping or as soldiers, although we also find a person designated as Chinese (“Jan de Chinees”), as well as a widow (Abigael, the widow of lieutenant Hendrick Hendrickx van Oldenburg), and another craftsman: a ship’s carpenter by the name of Dirck Janssen.Footnote ⁴⁴ From two contemporary maps we can glean that most of the plots (and the houses on them) in this block were fairly small, and hence presumably more affordable than larger plots elsewhere in the west; they were also conveniently located near the wharves and harbour, where several of the men listed here would have found work. If Cornelis had indeed departed from this house that morning, it would suggest he was on his way somewhere in a southeasterly direction, probably taking Spinhuisgracht or Jonkersgracht before turning onto Utrechtsestraat.Footnote ⁴⁵ We do not learn what Cornelis’s destination was, nor what purpose determined his route, but the movements of two others shed some light on the direction he took.

We can observe a similar movement oriented towards the southeast of the western inner city for a few others included in this case: the enslaved women owned by Thomé Svegedo. Svegedo, too, figures in the 1650 list of landowners. Listed here as Thomis de Sevedo, he owned two plots on opposite corners of a block is situated between the Spinhuisgracht, Utrechtsestraat, the Jonkersgracht, and the Maleischegracht.Footnote ⁴⁶ As Svegedo owned two plots, both home to properties of similar sizes and types, we are uncertain which of the two he occupied with his family and his enslaved servants. However, connecting these two locations and the spot where the women were attacked suggests a southeastward movement.Footnote ⁴⁷ A crucial explanation for this direction of movement is the small bridge over the Stadsbinnengracht, which gave people quick access to another, larger bridge that connected the western and eastern inner city.Footnote ⁴⁸ This would have been part of an attractive route for people living in the western part wanting to visit destinations in the eastern part (in which the church and City Hall were located) as well as the southern suburbs and Ommelanden beyond. Conversely, this route offered others moving in the opposite direction quick access to the shipyards and the market at the northern part of the western inner city as well as to public institutions such as the Spinhuis and the Chinese hospital. Therefore, it may not be coincidental that the other people on the move who became the attacker’s victims that morning were also gravitating towards this southeast corner of the western city.

Returning to Djago, we can infer from the above movements that she had cleverly positioned herself at the corner of Utrechtsestraat and the river adjacent to what in the report was referred to as the “Generaalstuin”, a plot of land owned by the Company.Footnote ⁴⁹ From the 1652 map we glean that this land consisted of well-organized functional gardens, fenced off and (presumably) for the use and benefit of Company officials only. Livestock might have been raised in its several circular and semi-circular enclosures; there were also rectangular plots with fencing, suggesting food production.Footnote ⁵⁰ Food shops and stalls, like Djago’s, are commonly positioned on streetcorners, as these are ideal places to attract customers coming from various directions.Footnote ⁵¹ In Batavia, too, tokos (shops) and warungs (street-food stalls) were also typically placed on streetcorners.Footnote ⁵² Her position on this particular corner facilitated the sale of food to passersby wanting to take advantage of the connection at Punt Diest between the western and eastern city.

Another option for crossing the river from west to east (or vice versa) was the barge that operated on the northern side near the Castle; recall that it was on this barge that our culprit claimed his first victim. It is possible that the use of this route may have been regarded as a less convenient option than crossing the river via the small bridge at Punt Diest. One was dependent on the barge’s speed and capacity, and when there was a lot of traffic one’s movement through the city would have been considerably slowed if one took the barge. Nevertheless, the barge connected the northern parts of the city and provided almost direct access to the Castle towards the northeast and to the fish market, warehouses, and wharves on the western riverbank. In 1655, only a few years after the case under consideration here, a small drawbridge was put up near the passage where the barge operated, indicating the need for a more permanent, and perhaps more reliable, solution catering to the needs of pedestrian traffic crossing the river at the northern side of the city.Footnote ⁵³

Probably attracted by a flow of traffic in this area similar to the place where Djago stood, the anonymous Chinese vendor had set up his pork stall on the corner between the river and the Maleischegracht (Malay canal). This canal separated a block of housing in the centre-east of the western city and the area that was home to one of Batavia’s iconic markets, famously depicted around the same time by the artist Andries Beeckman in his painting Gezicht van Batavia (View of Batavia, 1662) (Figure 3a). This market would have been open since seven in the morning, so our Chinese vendor was probably not the first to come and sell in this area. It is likely that he had visited his regular customers at their homes first, possibly to sell them his premium cuts, and then had moved to the area of the market to offer the remainder to marketgoers.

Source: Rijksmuseum, Amsterdam. Public domain.

Figure 3. (a) View of Batavia c. 1662 by Andries Beeckman. (b) Detail of Beeckman's painting, showing a Chinese meat vendor.

The Dagregister account gives little information on this Chinese salesman, but we do know he had carried baskets of meat (speck en vlees) and was cleaving the meat at his stall (craem) when he was attacked. Strikingly, in Beeckman’s painting, we find a male Chinese stallholder showing pieces of meat, probably pork, to a customer, based on his dress a Mardijker man, at roughly the same spot where the Chinese meat seller from our case was standing when the fugitive took his cutting knife (Figure 3b).Footnote ⁵⁴ In the painting, we see the Chinese vendor holding a knife between his teeth while a large knife, probably used for chopping cuts of meat, is lying on the surface of the stall, next to some cuts of pork, which he has put on display. Working from something that looks like a trestle table, he has baskets filled with meat cuts under his stall: you can just see, sticking out from under the table, the straps and poles that he had used to carry his baskets. Beeckman’s view of Batavia is generally viewed as a composition of types, all combined in one imaginary scene that served to display the cultural heterogeneity of the city. It was a view that was highly desired and admired, as evident in the painting’s being hung in the main meeting room of the VOC directors in Amsterdam and by references to the work in several contemporary accounts of the East India House by foreign visitors.Footnote ⁵⁵ Yet, since Beeckman is known to have made sketches in situ, and because he was in Batavia in the mid-1650s, it is likely that the appearance, tools, and working conditions of our anonymous Chinese pork vendor were not unlike what we see depicted in the famous painting.

Djago and the Chinese meat vendor are key examples of figures whose working lives generally remain obscured. Street vendors are often difficult to trace in archives, and their work practices are even harder to pin down.Footnote ⁵⁶ Women like Djago in particular often eluded registration, as their work activities were often not formalized, and they would have been less likely to have been included in formal registrations such as tax lists that, as we read above, favoured male heads of household. Similarly, the anonymous kaffer in charge of pulling the ferry (“over en weer trecken”) is an example of a type of worker who often escapes notice but who was nevertheless instrumental in the smooth running of the city. By placing these people within a space-time framework, we obtain better insight into not only their work practices but also the wider functioning of the city and its economy.

For Batavia, the case examined here is the first to be analysed in this manner, but for Amsterdam we have already collected a much wider set of cases, which yielded some striking results on gendered mobility and spatio-temporal patterns that to some extent resonate with what we see here. Based on some 1,400 observations on street use in Amsterdam across the eighteenth century, Bob Pierik has shown that, as in Batavia, urban mobility was highly dependent on the temporal regimes put in place by the urban government. The opening and especially the closing of the city gates, aligned with the activities of urban facilities such as markets and weigh houses, resulted in a remarkable shift in mobility.Footnote ⁵⁷ Figure 4 shows how far removed witnesses were from their homes at a particular time of day. It reveals a striking drop after the evening closure of gates at 9 p.m., suggesting that most people returned to their homes and neighbourhoods at precisely that time. From the Amsterdam data, it also becomes clear that, overall, men and women shared similar mobility patterns, but that women generally were found much closer to home.

Source: Pierik, 'Urban Life on the Move', p. 57.

Figure 4. The scope of mobility of men and women in eighteenth-century Amsterdam according to the time of day (distance from home in meters; N =1400).

This is an important finding, as women in early modern Dutch cities are generally portrayed as enjoying greater freedom of movement than their counterparts elsewhere in Europe. This finding is even more striking since many of the men and women who feature in these cases were from the lower to lower-middle strata. Earlier work by Robert Shoemaker on London revealed a gendered mobility pattern that suggested that for this social grouping male mobility was much more limited. Shoemaker found exceptionally high mobility among female servants, while male craftsmen were the least mobile group he encountered, as they worked out of the workshops attached to their homes.Footnote ⁵⁸ Our Batavia case, by contrast, contains both men and women at work away from home (the ferry operator, the Chinese salesman, the enslaved servants, possibly Djago), although it is worth pointing out that each of these workers was employed in the service sector, as were the highly mobile domestic servants in London. Yet, we also encounter members of both genders in the direct vicinity of their homes: a Chinese man and the wife of a Chinese shopkeeper; both the man and the woman were attacked in the area near Punt Diest. Women have been largely neglected in the above account, but we will now turn to them in order to explore what silences remain and how to remedy the problem of their obscurity.

Paper Realities: Silences, Vagueness, and What to Do about Them

In contrast to the individuals whom we met above, the information on activities, work-related or otherwise, of most of the other people involved in the case is limited. Of the two Chinese men who were stabbed, we know only that one of them, a man who went by the name of Siau Siecko, was standing on the doorstep of his house; the description of the other Chinese man, Kousiecko, gives no further information on his residence and reason for being where he was at the time of the incident. For the woman who held the child in her lap, it is recorded that she was married to a Chinese shopkeeper by the name of Banco. We may assume she was of Balinese descent, as many Chinese men in Batavia married women from Bali. Unlike men, women in China were not allowed to migrate abroad, and Chinese merchants who settled in Batavia regarded women from Bali as attractive marriage partners in part because they were not Muslim.Footnote ⁵⁹ It is not mentioned whether this woman was at or near her house when the perpetrator struck. The 1650 list of landowners does not contain any Chinese names for the area where she was attacked, so we cannot confirm her residency in this manner. If she did live there she was probably renting, as did Siau Siecko, who was reported to be at his house but was not listed as a landowner. Wives of small-business owners such as Banco are generally assumed to have been working in the same trade as their husbands, yet in-depth research into working couples has shown this was not necessarily the case.Footnote ⁶⁰ For Banco’s wife, the report in the Dagregister reveals no indication that she undertook activities in trade, but there is explicit mention of an important work task that is commonly overlooked in studies informed primarily by formal registrations of work: childcare.

A focus on childcare offers a crucial lens to further our understanding of gendered everyday work practices and urban mobility as well as a way to enhance our research practices, as it forces us to reflect on methods and on sources.Footnote ⁶¹ Along with a number of other types of care work, childcare was an important component of female work tasks, but occasionally also those of men.Footnote ⁶² In the case of Banco’s wife, we do not know whether the child was hers or someone else’s; if it was not her own child, she could have cared for the child as a favour, but also as a paid activity.Footnote ⁶³ The witness depositions we studied were extraordinarily silent on children, generally recording youngsters only when they were victims. Here, there were children in the midst of violence: the toddler mentioned above, as well as the under-five Chinese child who was playing in the street, a common activity, according to chronicler Valentyn, for the offspring of the Chinese and the enslaved.Footnote ⁶⁴ We also find the occasional child in our Amsterdam materials, typically recorded as victim: for example, children who are mentioned when neighbours launch a complaint about maternal neglect in the case of an impoverished family living in a basement dwelling.Footnote ⁶⁵ Children who may have been present while an argument unfolded or a crime was committed largely remained unregistered if unhurt. Since they were not adults, they were not seen as reliable witnesses, and no statements were taken from them. Their absence from the records means that childcare as a form of work also remains largely obscured.Footnote ⁶⁶

There is another reason, however, why information on childcare practices is lacking in these otherwise rich sources. While children generally did not narrate their experiences in the records, married women featured regularly in depositions, as they were commonly seen as reputable witnesses.Footnote ⁶⁷ Given that many of them were mothers, these witnesses could have provided an abundance of information on such care work; however, in most cases they remained silent about such practices. Even explicit searches for childcare in court records yield limited evidence.Footnote ⁶⁸ The reason for such paucity should be sought in a particular combination of factors: first, the mediation of the words of deponents by scribes, who as men of the law may have been less attuned to the various tasks involved in caring for children, and who in their roles as servants of the justice system had other priorities in their recording practices.Footnote ⁶⁹ Second, childcare comprised a mundane series of tasks that typically stretched over a longer period of time (throughout the day and night), characteristically combined with other work tasks that were prioritized in the statements.Footnote ⁷⁰ As historians of women’s work have shown for various contexts, motherhood did not preclude women taking on other roles, and many combined childcare (either of their own children or those of others, paid or unpaid) with other work.Footnote ⁷¹

Indeed, where these textual materials remain notably silent or at best vague on the tasks of childcare, visual materials on everyday city life show children and their carers in abundance. We find many depictions of women looking after children in genre paintings, for example in the domestic scenes that were so popular in the seventeenth-century Dutch Republic but also in representations of more public places, of the sort one finds in street scenes and church interiors. For example, in Emanuel de Witte’s market scenes, we observe women out shopping while holding children or see women portrayed in church, guiding a child by the hand or sitting down to breastfeed.Footnote ⁷² For Batavia, too, visual materials often portray aspects of childcare – most famously, the images included in the eighteenth-century diary of Jan Brandes depicting his son being cared for by one of their enslaved servants.Footnote ⁷³ But closer in time to our murder case we can find depictions of childcare in the sketches made by Wouter Schouten, who, around 1660, portrayed various aspects of everyday life in the city.Footnote ⁷⁴ Schouten depicted many a woman out and about accompanied by a small child. Crucially, several of these women were involved in another task while tending to the children. We see a woman holding a child by the hand while walking along a row of market vendors; another woman nurses a child as she watches a game of dice; and several women are depicted carrying their babies in slings while lifting heavy loads above their heads (Figures 5a–c).Footnote ⁷⁵ The questions that these depictions elicit are: What would these individuals have reported had they been captured not only in images, but also in their own words? Would they have testified they were carrying goods to the market, watching a game of dice, doing some shopping? Or would they have prioritized the care work, or mentioned both? Based on the above, it is likely that, in many cases, the former would have been foregrounded, illustrating the vital importance of incorporating visual materials into the study of gendered work practices.Footnote ⁷⁶

Source: Rijksmuseum, Amsterdam. Public domain.

Figure 5. (a) Street in Batavia with women and child in the foreground. Detail of Wouter Schouten, ‘Straat in Batavia met markt en een theatervoorstelling’ [Street in Batavia with market stalls and a theatre performance], c. 1660. (b) Woman breastfeeding. Detail of Wouter Schouten, ‘Twee groepen figuren’ [Two groups of figures]. (c) Women carrying and guiding children while transporting goods. Detail of Wouter Schouten, ‘Studieblad met lastdragers, dansers en vrouwen met manden op het hoofd’ [Study sheet with load carriers, dancers, and women with baskets on their heads], c.1660.

In their analyses, historians are often hesitant to treat visual materials as equal to textual sources. Apart from concerns about the tension between representation and reality, these materials have not often been used in more systematic and quantitative analyses. As Marie Yasunaga has illustrated for our Edo case study, however, turning pictorial qualitative evidence into quantitative evidence can be achieved by applying the same snapshot method we use to analyse witness depositions.Footnote ⁷⁷ As with textual evidence, for such analyses we ideally use serial sources that enable us to reasonably assume that they depict everyday practices in a fairly standardized manner. For Edo, she found such materials in the Edo Meisho Zue, an early nineteenth-century illustrated city guide showcasing Edo’s street life in seven volumes each containing some thirty to forty illustrations. These depictions are often highly detailed and are generally true to reality, as is evident in the striking similarities between the original sketches Yasunaga discovered and the printed versions of the images (Figure 6).Footnote ⁷⁸ What a preliminary analysis of this material reveals, among other things, is the commonality of childcare undertaken by women across various neighbourhoods in Edo. Most of these women carry children on their backs while hurrying along the street, buying provisions, or engaging in a conversation.Footnote ⁷⁹ The wide coverage and consistency of the materials allows us, through the application of the snapshot method, to reveal the spatial workings of Edo’s urban economy in a way that would have been difficult to achieve otherwise. Apart from spatial analyses, it also offers insights into the occurrence of combined (work) tasks, as well as into tasks such as care work that are commonly underreported in many textual sources.Footnote ⁸⁰

Source: Smithsonian Libraries and Archives, Washington DC, USA. Public domain.

Figure 6. Woman carrying a small child through an Edo street. Saitō Chōshū, Hasegawa Settan, Edo Meisho Zue, vol. 1, no. 49.

However, analysing visual materials in a more quantitative manner also presents challenges related to silences and vagueness inherent to these materials. While gender and social status can often be assigned through physical appearance and dress, the figures depicted are generally anonymous, bearing no names or lacking other personal information such as place of residence or family relationships; ages can be defined only via broader categories such as child, adult, or elderly person. The types of activities depicted must be deduced from actions, locations, and, for example, tools, but such activities can be identified, specified, and classified, for instance, through cross-referencing the image, as Yasunaga did, with contemporary maps of the city (Edo Kiriezu 1855) and listings of businesses (Edo Kaimono Hitorui Annai 1826). Labour relations can be inferred only occasionally, when servants and masters are depicted as participating in a single activity (for instance, in the case of palanquin carriers). However, for such types of information the processing and analysis of data require an additional layer of interpretation.Footnote ⁸¹ What our approach nevertheless illustrates is how, when word and image are combined, the understanding of work practices and the conditions under which they take place are greatly enhanced. For Amsterdam, the numerous women we found doing laundry on or near the canals depicted in eighteenth-century cityscapes confirmed that the single case we encountered in the notarial witness depositions was an example of what had been common practice at the time.Footnote ⁸² For Batavia, as we saw above, Beeckman’s painting revealed elements of the Chinese meat vendor’s work practices that remained obscure in the crime report (and vice versa). And when we turn to Djago, her work practices might have been similar to those depicted in the sketch of a woman by Wouter Schouten (Figure 7). This woman, identified as kafferin moffinjo in the accompanying text, is seated at what resembles a small portable stove, surrounded by a group of onlookers.Footnote ⁸³ Silences and vagueness in commonly used textual sources for the history of work, generally pertaining to work practices that are regarded as unremarkable, can thus be made audible and richer through the widening of our horizon. However, as our Edo case study shows, the materials that prove incredibly useful in this process – visual sources – are what deserve to be regarded as more than complementary materials. Like other forms of artistic representation, such as poetry, plays, and novels, visuals provide important glimpses into the everyday, and a systematic exploration such as the sort typically applied to texts offers crucial insights into work practices that are unrecorded or under-recorded elsewhere.

Source: Rijksmuseum, Amsterdam. Public domain.

Figure 7. Woman at a stove surrounded by onlookers. Detail of Wouter Schouten, Europeanen die naar een Indonesische man kijken en pijprokende figuren [Europeans looking at an Indonesian man, and pipe-smoking figures], c.1660.

The Clerk, the Artist, and the Historian: On Embracing Uncertainty to Achieve Better Data

In this final section, let us return to the question posed at the start of this article: How do we unite our searches for big and better data in a way that does justice to the complexities of historical societies and the sources that tend to capture only part of their essence? Our search for evidence on gendered street life and its myriad activities – both highly noticeable (fights or accidents) and utterly unremarkable (fleeting encounters or people passing the time) – has led to useful insights and subsequent research strategies that can be helpfully employed to enhance the quality of big data on everyday work practices. Layering data from different materials has proved crucial, as has applying a spatio-temporal framework that does justice to the complexity of city life. Our results also made clear how perceptions of what counts as an activity that is worthwhile to report varies significantly according to source type and context. This directly feeds into the issue of how work is defined and reported, and resonates with what scholars like Ågren, Whittle, and their teams have now been arguing for a while. As such, this may not sound like such a radical new insight. However, what we have encountered over the years is that historians appear to have different levels of tolerance for uncertainty depending on the character of the source material they employ.

While the tension between representation and reality is a constant presence in discussions of visual depictions of work, the question whether the incidental evidence presented in the witness depositions or in the diaries we use are truthful accounts of the past is often ignored. The reason for this lies in the differing assessments of the roles of artists and of bureaucrats, and the image versus the word. Most of us are well aware of the debates on source criticism regarding witness statements and diaries. We also acknowledge that, even if some evidence in such sources might be fabricated or can be seen as a form of storytelling, it nevertheless offers insights into what might be plausible or acceptable in a certain context. Fabrication and a certain level of creativity are thus tolerated for these written materials, whereas it seems more difficult to apply a similar attitude to visual materials, even when, as in our case, these depictions are very close to what an artist directly observed in the street.

However, the (perhaps heightened) sensitivity that we tend to apply to the image, and the role of its creator, should equally be applied to the written documents that narrate everyday experiences from the past. Just as artists may have omitted or changed aspects of the scenes they depict, early modern clerks or archivists may have done the same. As Frances Dolan and, more recently, Tim Stretton have argued, we must be careful in our assessment of what is said and by whom: Who is the creator of these documents? Who speaks? Who writes? How much might a writer know about the context of the events they are putting into writing? Is there a need to summarize (because of time or other constraints) and, if so, what is left out or described only in general terms?Footnote ⁸⁴ Asking these and many more questions along these lines is a crucial step in furthering our knowledge of the historical experiences of those who remained unheard and unseen, as well as of reinterpreting long-standing ideas about those groups and phenomena that have always been squarely in the limelight.

At a time when digital methods are fundamentally reshaping how we write history, there is both a warning and an opportunity in these observations. The computer’s preference for clearly defined entities and categories may exacerbate our blindness to certain phenomena and actors, while at the same time it is computing power and artificial intelligence that can help us to unravel where vagueness occurs, to illuminate what silences remain, and to discern how we can explain such patterns and as such best deal with the implications for history writing. While, in part, this might feel like a throwback to the insights and arguments of Microstoria and Alltagsgeschichte of the 1980s and 1990s, it is precisely this digital age that demands that we refamiliarize ourselves with their lessons and thus be able to push new boundaries in historical research. This call for a re-evaluation of microhistory and its employment in the history of work fits in a wider trend. In the last few years, several scholars of global history have advocated the integration of the microscale to understand global transformations and connections.Footnote ⁸⁵ They convincingly show that the macroscale of the global necessitates a return to the micro level; similarly, I argue that the macroscale of the digital also invites us to revisit microhistory. My emphasis here, however, is not on scale but rather on the way we read and interpret evidence. It appeals to questions of reading against the grain, of recognizing silences, and, above all, of embracing creativity and uncertainty, both inherent in the fabrication and the analysis of our sources as well as in the stories we subsequently tell.