Household Cohabitation Patterns in Multiethnic Seventeenth-Century Lviv

John Graunt, the founding father of demography, noted in 1662 that, in London, “there were about eight Persons in a Family, one with another, viz. the Man, and his Wife, three Children, and three Servants, or Lodgers.” Graunt’s picture of a London family (or household, as we would say today) probably reflected the social expectations of his middle bourgeoisie circles. Our current understanding of social diversity in premodern cities suggests that Graunt’s depiction oversimplified the great variety of domestic life in the English metropolis. Research on metric books has demonstrated that the mean household size was much lower than Graunt estimated and varied significantly based on a family’s wealth, housing conditions, and area of residence.¹

In the same year that Graunt published his seminal work on London’s demography, authorities in Lviv, a city about thirty times smaller than London and situated on the distant border of the Polish-Lithuanian Commonwealth, created a unique register documenting the city’s population. This poll tax register of 1662 offers a detailed description of the inhabitants of every burgher’s home within the city and its suburbs (excluding children up to ten years of age, Jews, and members of the clergy and nobility). Unlike ordinary municipal tax registers, it includes not only owners or main lodgers of a given household but also lodging families and all additional household members, including journeymen, workers, and domestic servants. It reveals almost complete living arrangements in each burgher’s house. Additionally, the 1662 poll tax register includes residents of suburban houses and several neighboring villages under the city’s jurisdiction. This historical document presents a unique opportunity to study different household cohabitation patterns in seventeenth-century Lviv and to explore possible explanations for these patterns.

Early modern Lviv had a distinct multiethnic and multireligious character. In addition to the politically and demographically dominant Catholics, Lviv was home to Orthodox Ruthenians (Ukrainians), Ashkenazi Jews, and Armenians. Each of these communities had their own communal authorities, legal privileges, and temples, as well as institutions such as schools, hospitals, and print houses. The particular social composition of the city followed the system of western urban law. Yet Lviv was also distinct from other western cities of its time as it played an important role as a trade center connecting Europe with points eastward owing to its location in the borderland. This location had given rise to an ethnic and religious mosaic of residents originating from diverse geographic and cultural spaces, giving historians an opportunity to consider the influence of structural and cultural features of space and ethnicity on domestic life in the city.²

The principal goal of this paper is to examine the size and composition of households in seventeenth-century Lviv and assess the influence of factors such as spatial location in the city, socio-economic status of heads of households, and ethno-religious diversity on urban domestic life. In combining information from the Lviv poll tax register of 1662 with additional sources such as regular Lviv tax registers and city plans, it is possible to determine the exact locations of nearly all of the houses inside the city walls and the size of the plots on which they were built and to investigate the possible determinants influencing household cohabitation patterns.

A growing number of scholars emphasize that social processes are not solely shaped by economic, legal, and cultural conditions, but also by the qualities of the space in which they occur. As Arnade, Howell, and Simons highlight, space cannot be reduced to either “a bricks-and-mortar inertness,” nor can it be treated purely as a discursive phenomenon. Instead, space should be understood as “a material site that both contains and generates cultural references, economic and political meaning, and social forms.”³

At the same time, these meanings are contingent upon the material features of objects in space that provide shelter and space for work and rest, such as the size, shape, and quality of workmanship. Space is therefore not merely a setting or a stage of human interaction, but the very site of resources and the constraints affecting them. Wirth argues that the place of life determines the way of life. This aspect of relation between the urban space, its social meanings, and domestic life has been largely overlooked in previous literature, though there is a growing recognition of microanalytical approaches linking places of residence to forms of domestic life.⁴

Scholars of family life have observed two distinct features of premodern urban households. First, these households exhibited complex structures composed of people who were unrelated to the head of the household. Second, multiple households coexisted within a single house. These phenomena were rarely found in the countryside, but, as Merry and Baker have observed, they were common in cities, especially large ones. Late-seventeenth-century sources from London record instances of multi-family cohabitation within households, as well as a high incidence of lodgers, apprentices, and servants alongside a relatively small number of children. Lindström confirmed these observations in his study of the eighteenth-century Swedish town Linköping, underlining the importance of analyzing households within a wider housing context and considering the “spatial dimensions of social organization of household structures.” He referred to this aspect of relation between space and family life as research on “cohabitation patterns.”⁵

Lviv was founded by Daniel of Galicia, a Ruthenian ruler from the Rurik dynasty, in the mid-thirteenth century. After its conquest by the Kingdom of Poland in the mid-fourteenth century, Lviv underwent significant changes. King Casimir III enacted the Magdeburg Law system, abolishing the subordination of Lviv’s inhabitants to Ruthenian law. The city’s domain was redefined, and its economic independence was supported by a royal gift comprising the large surrounding territory. Accurately estimating the size of the city’s population during this period is difficult, but it is believed that during the Middle Ages the city and its suburbs had about 5,500 inhabitants. By the beginning of the seventeenth century, the population had grown to approximately 14,000, reaching its peak in the early modern period. Of these inhabitants, about 4,500 lived inside the city walls.⁶

The area of the city within its medieval walls was a mere 21 hectares. In this space in the seventeenth century, there were about 330 residential houses, several monasteries, and various economic, communal, and defensive buildings that also housed permanent residents. Lviv’s layout resembled that of other cities in central and eastern Europe in the late Middle Ages. The city center was the town square, in the middle of which were merchant stalls and the town hall, seat of the municipal authorities. A parish church—later a Latin cathedral—was built on the corner of the square, and the main road passed through the square, guarded by Krakow and Halich Gates. These gates were part of the city’s fortifications, having been gradually expanded since the Middle Ages. They consisted of twenty-five defensive towers, the city walls, moats, ramparts, and the Low Castle, which had its own fortifications. Atop a hill near the city stood the High Castle, a former center of princely power. The territory around the hill was also inhabited, albeit outside the legal jurisdiction of the municipality. In front of the city gates were the two main suburban settlements, the Krakow suburb to the north and the Halich suburb to the south.⁷

Following the enactment of the Magdeburg Law, a new symmetrical city plan, a common feature among cities in the region, was marked out. The plan was characterized by a demarcation of distinctive residential plots and a system of streets emerging from the market square and intersecting at right angles. Larger plots were created around the market square and smaller ones were relegated to side and back streets. This functional and organized urban plan was embedded in the ideological system of Magdeburg Law. It strengthened the unity of the community, upheld the authority of self-governance, reflected the social hierarchy of its inhabitants, included the religious sphere, and separated the new community from the outside world with a city wall.

In the case of Lviv, however, the implementation of this orderly plan faced challenges from the beginning. In the early fourteenth century, Lviv was already inhabited by merchants and craftsmen from distant lands who served the Ruthenian princes. Consequently, following Casimir III’s conquest of the territory and the privilege he granted in 1356, other ethnic groups (known as gentibus) coexisted alongside the Catholic community. These groups included Armenians, Jews, Ruthenians, and Tatars. The king gave them the right to use the common Magdeburg Law or to follow their own distinct ethnic laws in the presence of the royal superior of the city, known as the vogt.⁸

Members of these ethno-religious communities resided inside the city walls where they built their houses and established places of worship. Their lower social status compared to the dominant Catholic community was emphasized, in part, through territorial segregation. These groups settled in the smaller spaces of Lviv, situated along the back streets running along the city walls. Over time, these areas received designations in the form of ethnic street names, such as Armenian Street, Ruthenian Street, and Jewish Street (the Tatar community had disappeared over time). The process that shaped the character and boundaries of these spaces remains inadequately researched, but it can be assumed that these areas gradually developed into unambiguous ethnic districts. Historical accounts from the sixteenth and seventeenth centuries attest to the townspeople’s mutual efforts to maintain ethno-religious homogeneity among owners of individual plots.⁹

The latter half of the seventeenth century is regarded as a period of deepening crisis in Lviv and the broader Polish-Lithuanian Commonwealth. Between 1648 and 1667, the country was mired in continuous wars that severely strained the state’s power, led to the devastation of large parts of its territory, and weakened the economic position of the urban population. Lviv itself was besieged by hostile troops in 1648 and 1655, first by Cossack-Tatar forces and then by Cossack-Moscovian forces. The city managed to defend itself at the steep cost of a significant ransom, periodic famines, and intensification of the plague, typhus, and smallpox epidemics between 1648 and 1663.

The suburbs of Lviv also suffered serious damage. In 1638, fires destroyed houses in the Halich suburb and two years later in the Krakow suburb. Reconstruction of those areas was interrupted in 1648 and 1655 when townspeople deliberately set fire to houses to prevent them from being used as shelter by advancing enemy troops. Nevertheless, the reconstruction of the suburbs must have taken place relatively quickly, because from 1668 there is evidence of the existence of 736 buildings, including manors, wooden houses, and cottages in the city’s suburbs and villages. This fact suggests a substantial demographic growth, despite the significant losses the city had suffered.¹⁰

This study relies primarily on the poll tax register of 1662, which is the earliest source documenting the majority of Lviv’s population in the early modern period and one of the earliest urban micro-censuses from the Polish-Lithuanian Commonwealth. Modern state-ordered censuses became more common only in the second half of the eighteenth century and so have survived in a greater number. Supplementary analyses were conducted using the city tax registers, including the szos taxes of 1662 and 1663, the war contribution register of 1655, and hearth tax register of 1668.¹¹

The Lviv poll tax register of 1662 was commissioned by the Parliament of the Polish-Lithuanian Commonwealth with the purpose of paying off rebellious troops. It aimed to include all residents of the country, regardless of their social status, sex, and nationality (excluding children under the age of ten, resulting in their exclusion from the register). Furthermore, municipal authorities were responsible for tax collection, but the nobility and clergy paid taxes independently to maintain social distinction and emphasize their legal independence. Jews also paid taxes separately in the form of lump sums to the royal treasury. As a result, these groups were not included in the poll tax register.¹²

Although the poll tax was to be collected from “every head,” the amount due depended, in part, on the citizen’s place of residence, wealth, occupation, and marital status. Regular residents of larger cities, including Lviv, were required to pay 2 florins each, while inhabitants of the smallest cities, suburbs, and villages owed 1 florin each. Those holding important state or city offices and those engaged in financially profitable ventures were obliged to pay more taxes. The representatives of the highest municipal offices—mayors, councilors, and court mayors—paid an additional 10 florins each. Owners of printing houses and merchants involved in lucrative trade (such as silk, cloth, wine, spices, and grain) were to pay an additional 15 or 30 florins, depending on their wealth. In practice, the trade surcharge varied widely, ranging from 2 to 30 florins, indicating that it was based on an estimate of the value of trade commodities. Similarly, the poor were exempted from paying even the basic tax rate.¹³

The 1662 Lviv register contained information about different types of buildings, including residential buildings, special-purpose buildings, and defense buildings. The residential buildings included brick houses, small brick houses, wooden houses, and small wooden houses. Special-purpose buildings included church institutions, such as monasteries, hospitals, and schools, and economic or industrial buildings, which also served as residences. Defense buildings included city fortifications, which were partially inhabited when the register was made.

The inhabitants of the buildings were listed in a strictly defined order that reflected the social hierarchy of the city and the distinctions within individual households. The records used a three-tier division of family, employees, and servants. This class system was closely intertwined with a gender division. The names of wives were listed only after their husbands, daughters were usually named after sons, and female servants after male ones. Although there were exceptions, this general rule dominated the registry.

However, there were slight variations in the recording format for taxpayers within the city walls and for those outside. In the city, the name of the building included information about its current or previous owner and building type (for example, “Mr. Szulc’s brick house”). The buildings on several main suburban streets, on the other hand, had simple numerical names (such as “first house”). The houses of most secular and clerical suburban land estates and villages were not mentioned, and their inhabitants were simply listed one after the other while maintaining clear divisions between individual households. This suggests a decrease in accuracy farther from the city center toward the peripheries, where lower-status inhabitants resided. Altogether, the register listed 175 private houses, 33 farms and larger private estates, 12 mansions and manor houses, 7 urban villages, 1 hospital, 1 mill, and 1 stable in the suburbs.

To verify the completeness and representativeness of the poll tax register, and to supplement information about the size of the city, this study also uses data from the szos tax registers of 1662 and 1663. The szos tax was an elementary form of urban tax that was applied to cities in the Polish-Lithuanian Commonwealth. It had been collected since the Middle Ages, usually once or twice a year, for municipal or royal needs. In the second half of the seventeenth century, it comprised two taxes, the land tax paid by property owners and the property tax collected from the heads of lodging households. The amount of land tax was determined by the size of the municipal plot and corresponded to one’s share in the property. The amount of property tax levied on the heads of lodging families was based on the estimated value of the property.¹⁴

Although the basic unit of urban residential territory was initially an area (approximately 900–1000 m²), in the seventeenth century, due to increasing fragmentation of urban property, Lviv’s authorities began using the chlop (equivalent to 1/4 of an area). The sizes of the town plots ranged between 0.5 and 6 chlop, with 1 or 2 chlop being the most common sizes (around 250 or 500 m²).

There are slight differences between the lists of houses in the 1662 poll tax register and those in the szos registers. The szos registers included seventeen buildings that were not listed in the poll tax register. Because Jews and clergy were excluded from the latter, it is likely that some of these buildings were owned by religious institutions or inhabited only by Jews. Similarly, the poll tax register mentioned twenty-five inhabited locations in the city that did not appear in the szos registers. Among these were religious institutions like the Catholic Hospital of St. Spirit, the Ruthenian school, and the Orthodox monastery of St. Onuphrius. Additionally, there were buildings intended for municipal employees, such as the doorman’s house and the clockmaker’s house, as well as industrial municipal buildings such as the Shears, the Weight, and the city stable.

The szos registers also omitted numerous inhabited fortifications, such as towers and gates, as well as small trade establishments like butchers and booths. These places, although located within the city walls, were not subject to szos taxation because they were not part of residential plots. According to the 1662 poll tax register, these areas were inhabited by 269 individuals, mostly from lower-income groups.

The collected data have significant limitations that must be considered if we are to draw broad conclusions from them. The first limitation pertains to domestic space. Although we can locate houses on the city map, establish the size of their plots, and collect information about the number of families living in them, the data are insufficient to understand the type, size, and quality of space occupied by each house. This limitation prevents us from exploring the arrangements of families within multi-family houses, how many rooms a family was entitled to, and where these rooms were located.

The second limitation is the absence of information about the ages of the inhabitants, which severely restricts the ability to draw conclusions regarding the influence of the life cycle on family, work, and housing conditions. Third, it is not possible to verify precisely the ethnicity and religion for all individuals in the register. To include ethnicity in the analysis, each residence in ethnic districts was assumed to correspond to a particular ethnicity, although there were exceptions to this generalization. Finally, the possibility of comparing living conditions inside and outside the city walls is limited due to the disparate sets of information contained in the tax registers. Despite this limitation, we supplemented information about the suburbs wherever possible.

The heterogeneity of Lviv’s territory can be seen clearly when the city is divided into distinct areas (Fig. 1). This partitioning is based on the conventional concentric plan of the city, with a central wealthier area surrounded by poorer peripheries, while also delineating ethnic districts within the city.

The first area comprises large brick houses on all four sides of the central market square. Smaller brick houses are in the center of the market square next to the town hall and merchant stalls. Moving outward, the third area, hereinafter “I Cross,” consists of houses on the streets emanating from the market square, including Krakow, Halich, and five smaller streets. The next three areas, hereinafter “II Cross,” encompass the back streets. The II Cross includes ethno-religious districts and is subdivided into the Armenian quarter, the Ruthenian quarter, the Jewish quarter, and the II Cross Catholic houses. Figure 1 also illustrates the locations of sacral buildings, ecclesiastical property, and the outskirts, which include the houses on the rampart behind the Krakow Gate and the town fortifications in which some families lived.

This analysis compares these seven areas (the sides of the market square, the center of the market square, the I Cross, the Catholic II Cross, the Armenian quarter, the Ruthenian quarter, and, wherever possible, the city outskirts), and tests the assumption that these areas are characterized both by their relative internal similarities and by their visible differences in terms of the size of their plots, their economic value, and the social composition of their inhabitants (see also Table A1).

The analysis excludes data on the inhabited municipal industrial and economic buildings, such as stables, foundries, stalls, and booths, as well as buildings belonging to religious institutions, such as schools, hospitals, and monasteries. This exclusion is due to the lack of information on the size of residential space in those buildings and the assumption that the composition of their inhabitants differed significantly from that of inhabitants of residential houses. Similarly, this study also omits analysis of the Jewish quarter due to a lack of data.

Figure 2 illustrates the key characteristics of Lviv’s neighborhoods in terms of their size, number of inhabitants, the households those inhabitants belonged to, and the houses in which they lived. The city can be divided into three main parts—the market square, the I Cross, and the II Cross, which includes ethno-religious quarters. These districts account for approximately equal portions of the residential urban space. The distribution of households and inhabitants among these major areas is also similar; this was due to a similar density throughout the whole city. On average, each chlop (about 225–250 m²) accommodated between six and eight people and between 1.4 and 1.7 households.¹⁵

Yet there were significant differences of size and population within the ethno-religious districts. For example, the Ruthenian district was territorially four times smaller and over three times less populated than the Armenian quarter and the Catholic part of the II Cross. The size and population of the Ruthenian district closely resembled that of the small brick houses in the market square.¹⁶

The relationship between the number of houses and size of plots indicates significant differences in the average plot size across the city. The houses around the market square were situated on plots that were, on average, twice as large as those on the back streets of the II Cross. Not only were the market square houses larger, but the plots had the capacity to accommodate additional buildings, many of which were also inhabited.

Cities are often characterized by the coexistence of extreme poverty and enormous wealth within a relatively small area. However, urban sources rarely provide detailed depictions of poverty, often presenting it as a nameless mass rather than individuals with their own families and specific living arrangements. To better illustrate the diversity of Lviv’s population and its economic inequalities and the resulting impact on urban cohabitation patterns, the heads of households have been grouped into four categories based on the amount they were obliged to pay in taxes.

The first group comprises the wealthiest individuals, including the richest merchants and city authorities, who paid between 12 and 160 florins each. They account for 9 percent of all heads of families. The second category includes those who paid a considerable tax ranging from 2.5 to 10 florins, representing 11 percent of all heads of families. This group included members of smaller trades, pharmacists, and the richest craftsmen. The third group consists of taxpayers who paid the standard tax of 2 florins, or 1.5 florins for female heads of families. This group, which was mostly composed of affluent craftsmen, constituted 40 percent of all heads of families. The last group includes payers of reduced taxes ranging between 0.5 and 1 florin, representing 39% of all heads of families listed in the register. Half of these families were described in the register as single-person households, though some of them may have included young children under 10 years old.

Comparing the social composition of Lviv’s neighborhoods reveals not only the expected differences between districts but also their internal diversity (Fig. 3). Taxpayers from the first and second groups, which together constituted 18 percent of all heads of families, accounted for 58 percent of households in the small houses in the market square, 38 percent in the Ruthenian quarter, 30 percent in the big brick houses around the market square, and 29 percent in the Armenian district. They were almost absent from houses on the back streets of the Catholic II Cross and the outskirts of the city. At the same time, the urban poor, comprising about 30 percent of households, were prevalent in some areas and relatively scarce in others. In the market square, they constituted only 10 percent of the population. Conversely, they formed the majority in the city peripheries, accounting for 63 percent in the Catholic II Cross and 77 percent in the outskirts. The houses located on the streets of the I Cross had a distinct character, as they were inhabited predominantly by wealthy craftsmen from the third tax group.

A different tax scale was used for residents living outside the city walls (Fig. 4). The standard tax in the suburbs was 1 florin, but actual paid taxes ranged between 0.2 and 2 florins, allowing the categorization of taxpayers into three groups: those who paid more than 1 florin, those who paid 1 florin, and those who were eligible to pay less than 1 florin. Although the tax scale in the suburbs does not reflect significant differences in the wealth and social positions of inhabitants, it provides insights into certain economic disparities among the various suburban estates. Notably, the private houses on the main suburban streets stand out, with 40 percent of heads of families paying a higher tax rate there, while manor farms had the largest percentage of those with partial tax exemption.

A feature of urban domestic life was the inhabitation of individual houses by multiple households sharing common spaces, such as entrances, yards, staircases, and sometimes kitchens and washing facilities. The houses around the market square, situated on larger plots, were inhabited by an average of twenty people each, comprising five separate households per building. These houses were inhabited by rich merchants with their families and servants, wealthy craftsmen who sublet chambers for themselves and their apprentices, and widows and poor craftsmen who lived behind the houses in cellars and wooden outbuildings.¹⁷

Farther from the city center, both the size of plots and number of inhabitants decreased. Houses on the side streets of the I Cross were inhabited by an average of twelve people each, forming three households, and on the back streets of the II Cross (including the Armenian and Ruthenian quarters), houses had an average of six or seven inhabitants each, comprising two households. In the wooden houses on the embankment and suburban streets, these numbers were slightly lower (Figs. 5 and 6).

Family history researchers have long emphasized the household as the basic unit of social life in the premodern period. It was the main site of production, consumption, and socialization. Its members shared a common “roof and table,” and although only some were related, the rest were contractually part of the same household. The size of a household depended not only on individual preferences and needs, but also on the abilities of the heads of families to provide shelter and food and to pay the agreed remuneration for employees and services.

It is therefore no surprise that the differences in average wealth and residential space in the different areas of the city were reflected in the size and structure of households. The highest average household sizes were seen in the center of the market square, with five people per household, followed by 4.8 people per household around the market square, approximately 4.5 in the I Cross and Armenian quarter, and 3.6 in houses in the Ruthenian district and the Catholic II Cross. The average falls even lower in the city’s outskirts and suburbs, at about 2.5 to 3 people per household.

Figure 7 shows the differences in the mean household sizes between the taxpayers’ groups in the city and the suburbs. The highest tax group had an average household size of six people, while the second and third tax groups averaged four people. The poorest group had the smallest households with an average of two people. In the suburbs, households paying the higher tax rate averaged only three people, those paying the standard rate averaged two people, and the majority of those paying less than 1 florin were single.

A regression analysis tests the statistical significance and influence of location within the city on the size of the household while accounting for the socio-economic situations of household heads. Figure 8 and Table A2 present a model in which the dependent variable is the household size. The main determinants of household size in the model are the area of the city (where I Cross is a reference category), tax group (the third tax group is a reference category), and the size of the plot. Control variables included the household number in the register (coded as 1 for the first listed household, or the house’s owners or main tenants and 0 if otherwise), gender of the household head (coded as 1 for male and 0 for female), and citizenship status (coded as 1 if the head of household was titled in the register as “Sir” or “Madame” and 0 if otherwise).

This analysis shows the significant relation of the location within the city to the size of the households, even after controlling for individual wealth, plot size, owner/lodger status, city citizenship, and gender. Households in the II Cross exhibited significantly smaller sizes than those in the I Cross, a trait evident across all ethno-religious districts of the II Cross. Moreover, household size was related to household wealth; belonging to the lowest tax group had a negative effect on household size, while belonging to the highest tax group had a statistically significant positive effect. As expected, the status of the owner or main tenant, their gender, and citizenship designation also had significant positive effects on household size.

Lynch claimed that there was a characteristic demographic regime in the cities and countryside of northwestern Europe. This pattern—described by Hajnal as the European Marriage Pattern—is characterized by the prevalence of nuclear families, relatively late marriage ages (over twenty-three for women and over twenty-six for men), high levels of permanent celibacy among city inhabitants, and the practice of establishing new, separate households after marriage, known as neolocality. Lynch argued that this phenomenon extended beyond large mercantile centers like London and Paris to smaller towns in Germany and Switzerland (called “home towns” by Walker). The features of the phenomenon were conditioned by both the socio-economic relations within urban institutions such as craft guilds and in a sustained and widespread system of cultural values. Recent studies have shown that the main features of the European Marriage Pattern are traceable beyond the Hajnal line and also occur in the larger cities of the Polish-Lithuanian Commonwealth.¹⁸

When analyzing households, it is important to consider not only their size but also their composition, specifically the social and family roles of their members. In this study, households are classified into four types based on the basic schema proposed by Laslett—the individual, single-person household; the individual, single-person household with servants; the simple family household, which consisted of nuclear families (a married couple with offspring or without, and also widowed person with offspring); and the extended family household, which consisted of a head of household and one or more relatives other than spouse or offspring. Because the Lviv register does not include children under ten, the first type encompasses both single-person households and potentially widowed individuals with children under ten. Laslett also distinguishes multiple and joint family types composed of two or more married couples; these were rare in the Lviv register and are considered part of the extended family category.¹⁹

Figure 9 presents the distribution of family types across different areas of Lviv. The most common family type in the city was the nuclear family, accounting for 54 percent of all families in the city and 72 percent in the suburbs. Notably, in the prestigious houses around the market square, nuclear families constituted only 22 percent of all households, largely due to the prevalence of single-person households in this area.

Extended families constituted 9 percent of households within the city and about 5 percent in the suburbs. The limited presence of extended families in urban communities is consistent not only in the seventeenth-century poll tax registers of Lviv and Warsaw, but also in London and many larger and smaller towns of the Polish-Lithuania Commonwealth at the end of the eighteenth century. In Lviv, the majority of extended households included one or sometimes two female relatives, such as mothers, mothers-in-law, sisters, nieces, or granddaughters. Households with male relatives or a greater number of relatives living together were much less common. Although extended households appear in all the areas of the city and its suburbs, they are noticeably more common in the Armenian quarter (23 percent), the small houses on the market square (19 percent), and the big houses around the market square (16 percent)—areas also inhabited by a large number of merchants. Extended families were also more common among the two wealthiest taxpayer groups and considerably less common in the lowest taxpayer group.²⁰

Single-person households with servants constituted about 10 percent of all households in the city and only 3 percent in the suburbs, a proportion similar to that observed in late seventeenth-century London. An important feature of these households was the predominance of female heads; within the city walls, 68 percent of single-person households with servants were headed by women, and in the suburbs, this number rose to 78 percent. It is likely that most of this group were widows or widowers who had not remarried. Such households can be found in all seven areas of the city and across all tax groups, although they were most often in the lowest taxpayer group.²¹

Lone householders constituted a significant part of all households (26 percent in the city and 20 percent in the suburbs). These figures are also seen in early modern Poznań (21 percent) and London (34–37 percent depending on the district). It should be noted, however, that the category of single-person households may also include widows with children under ten years of age. Notably, houses around the Lviv market square had a high concentration of single-person households, constituting over 50 percent of all household types in that area. These households were primarily poor; in the city, 83 percent of them belonged to the lowest tax group, and in the suburbs, most of them (71 percent) were among the payers of the basic rate of 1 florin. Women were the majority within single-person households, representing 65 percent in the city and 63 percent in the suburbs. It is sometimes difficult to determine whether these people were indeed heads of single-person households, part of larger households, or were servants in some other house.²²

Only 311 children were mentioned in the register as living in the city (9 percent of inhabitants) and 139 in the suburbs (about 4 percent). Because children under the age of ten were exempted from the poll tax, the listed children were likely adolescents or adults living with their parents. Moreover, historical studies indicate that young individuals often worked as servants or apprentices in other households as part of their development and education, starting as early as the age of ten. It is therefore possible that young inhabitants of Lviv could be recorded in the register as servants, which could be a partial explanation for the small total number of children.²³

The average number of children living with their parents, as recorded in the Lviv register, was compared to data from the poll tax registers of Poznań in 1590 and Warsaw in 1659, which included children of all ages. Data from seventeenth-century London households was also included for reference. Although each of the registers has its limitations, this comparison provides a framework for the mean number of children living with their parents in an early modern city. The shortage of children in the Lviv register is evident, as it lists fewer than half the number of children and households with children compared to other registers (Table 1), which suggests that about 500–600 young children were not listed in the Lviv register.

The deficiencies in recording children in the suburbs are probably even greater. Direct comparisons with the city center are difficult due to the distinct social structure of the suburbs, including factors such as lower wealth, a higher proportion of migrants, and possibly different mean ages among the inhabitants of the urban space inside and outside the city walls.

Interestingly, the 1662 register documents, on average, more households with daughters than with sons in families across all areas. The only exception to this trend was in the Ruthenian quarter (Fig. 10). In most areas of the city, about 20–30 percent of households of two or more members had adolescent or adult children living with their parents. Yet in the small houses on the market square and in the Armenian district, these proportions were much higher, around 40 percent. This might partially be explained by the prevalence of wealthy Lviv merchants in these areas who could afford to provide care and education in their own homes.

Figure 11 shows the distribution of the mean number of children over the age of ten across the four groups of taxpayers. The result confirms that the average number of children increased with personal wealth. In the wealthiest taxpayer group, 41 percent of all households of two or more members had children, whereas in the lowest taxpayer group, only 23 percent of such households had children. Interestingly, the proportion of daughters among the children followed an opposite trend. In the richer groups, girls accounted for about half of all children but constituted 60 percent and 85 percent of all children in the two lowest taxpayer groups, respectively.

This phenomenon suggests two possible reasons why children stayed at their parents’ homes. In affluent urban society, a relatively high percentage of children of both sexes lived with their parents to help them, receive home education, and learn a trade before getting married and starting their own households. In poorer homes, the financial burden of raising children was considerable. Sons were often sent to other families to learn a trade or provide service, and daughters remained under their parents’ care for longer periods, providing help in lieu of hiring servants, possibly due to the lack of sufficient dowries. Moreover, among less wealthy burghers, learning a trade and serving in another home as a “life-cycle servant” was considered important for achieving independence and reaching maturity.²⁴

In seventeenth-century Lviv, 75 percent of households within the city walls of two or more members employed, on average, 1.9 servants and hired workers per family. In the suburbs, the number of households with servants was significantly lower (40 percent). The head of the household provided servants with food and shelter in exchange for their time and work, sometimes learning a craft as pupils or apprentices. There was a hierarchy among servants based on their role, age, gender, and remuneration, and poll tax collectors used a specific system to record household members. Male employees, including journeymen and apprentices, were mentioned first, then students, boys, farmhands, and coachmen, followed usually by female workers such as innkeepers and vendors, cooks, housewives, maids, girls, and nurses (Fig. 12).

The 1662 register records the largest group of male workers in the city as journeymen, followed by “boys” and farmhands. Most women were cooks, followed by the broad category of “girls” (named in the source as dziewczyna, dziewka, niewiasta, białogłowa, do usług, służebna), and then maids. In the suburbs, the number of male employees was notably lower. The majority of female servants in the suburbs were unskilled girls, followed by cooks (Fig. 12).

The number of servants in households varied greatly, ranging from zero to eleven in one family. Their average number depended significantly on the location of residence and the financial status of the heads of households. The houses around the market square, small houses in the market square, and houses of I Cross had an average of 2.1–2.4 servants per household. In all three areas of the II Cross, the average was 1.4–1.6. In the outskirts, the average number of servants per household fell below one. Wealthier houses on the main suburban streets had an average of 1.2 per family. Servants were rare the rest of the suburbs.

Most households had only one or two servants (61 percent of all cases in the city), making it necessary to interpret their specific roles with caution. Male workers were largely the productive forces in workshops or merchant enterprises, while women more often performed work related to food and the comfort of domestic life.

Figure 13 shows the distribution of male and female servants across the city districts and the outskirts. The highest concentration of servants was in the houses around the market square and in the I Cross. Across most of the city and the suburbs, the ratio of male to female servants was fairly balanced, but in the small houses in the market square, as well as the Armenian and Ruthenian districts, female servants were more common, possibly due the prevalence of other types of professional activities such as trade dominated by the work of family members or salaried workers from outside the household.

The average number of servants was strongly related to wealth and profession. The wealthiest group of taxpayers had an average of nearly four servants per household, while in the second and third taxpayer groups there were about two servants per household, and the urban poor had an average of only 0.7 servants per household.

The distribution of servant categories cannot be solely attributed to household wealth. Although there is a general decrease in the number of servants in households with lower levels of wealth (Fig. 14), the number of apprentices and pupils was highest in the third tax group. Up to four apprentices and three pupils worked in workshops at a given time, which can be attributed to the traditional guild system, where master craftsmen would take in apprentices and pupils to work and study in their households. Wealthy craftsmen’s families could afford to support several such unrelated household members. Less affluent traders relied more on family members and short-term hired assistance. In the poorest tax group, it was still possible to find some apprentices, but female service was practically non-existent.

This article highlights the influence of social inequalities, including economic, spatial, and ethno-cultural factors, on the diverse cohabitation patterns in Lviv. Merry and Baker, who analyzed domestic life in two different areas of London in 1695, stated that “the households occupying these houses were clearly influenced by the physical environments that they occupied, their character and composition being shaped by the housing stock, its size, quality, value and availability.” The research presented in this paper not only confirms those observations, but strengthens them by providing solid, data-driven evidence from a different geographical and cultural region.²⁵

Although the population density was even throughout the city, there were significant differences in living conditions in its different areas. One of the crucial aspects of spatial heterogeneity in Lviv’s urban territory was a diversity of residential plot sizes. The largest plots were in the city’s center around the market square, and their average size diminished with increasing distance from the center into the city’s peripheries. This structural aspect of urban space influenced the phenomenon of families cohabitating under “one roof.” Houses on the back streets were usually inhabited by two small households living in the same building, the main streets had an average of three families per house, and the market square had five families per house. This resulted in higher socio-economic diversity in the city center compared to the back streets. Additionally, living conditions were likely also more varied in the center, with inhabited basements, attics, utility rooms, and outbuildings. On the other hand, small houses in the market square (inhabited by wealthy merchants) and those in the poorer areas near the city walls had more homogeneous populations.

This research reveals that differences related to place of residence and ethnic identity, in addition to socio-economic factors, played a key role in the variability of domestic life in premodern Lviv. Comparative analyses highlight significant differences between the city’s seven areas in terms of household structure, wealth, and size. The city exhibited a traditional concentric urban pattern, with wealthy merchants with the highest number of servants residing in the center, and smaller households on the back streets, corresponding to lower socio-economic status. The Armenian and Ruthenian quarters deviate from this pattern due to the presence of wealthy merchants. Both ethnic districts show specific characteristics related to family types, number of children, and types of servants. Although socio-economic factors had an impact on the conditions of domestic cohabitation, the relationship was not strictly linear. City citizenship, house ownership, and living in “better” areas all influenced the average household size and the number and gender of children at home. Similarly, one’s profession influenced the type and number of servants, with merchants requiring different types of workers than stallholders, vendors, or artisans. Our most important conclusion, though, is that even when controlling for socio-economic status, the distinct areas of the city significantly impacted household size.

These findings show a large variation in the size and composition of households in seventeenth-century Lviv and in the conditions of their cohabitation within individual houses. These differences are influenced by economic, spatial, and ethno-religious factors that have received limited attention in the existing literature. Future studies should move toward a more comprehensive understanding of housing conditions in cities, both in residential and non-residential buildings, and analyze the transformation of private houses into multi-family rental dwellings, a common practice in large cities at that time. Such investigations are important for deepening our knowledge about urban family life in the premodern era.